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: 5.0

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
27.9249	0.0658	5	36.1233	0.6096	0.0235	0.4366	0.1591	0.5323	0.1780	0.0605
9.1858	0.1316	10	12.7427	0.6227	0.9763	0.4648	0.1364	0.5591	0.6550	0.5624
0.0007	0.1974	15	29.0077	0.5875	1.0000	0.7958	0.5909	0.7043	0.9997	0.9776
39.3778	0.2632	20	10.1170	0.6285	0.9997	0.3380	0.0455	0.4839	0.9328	0.9227
20.1992	0.3289	25	7.1120	0.6236	0.9555	0.4789	0.1818	0.5591	0.7503	0.6556
7.1302	0.3947	30	7.8105	0.7222	0.5312	0.6549	0.2727	0.6720	0.6232	0.3266
23.168	0.4605	35	7.6526	0.7580	0.9955	0.6901	0.2273	0.7097	0.9673	0.8339
4.4026	0.5263	40	5.1725	0.7995	0.8929	0.8592	0.2955	0.8226	0.9443	0.6850
10.4761	0.5921	45	4.4398	0.7766	0.6261	0.7746	0.25	0.7688	0.7958	0.4420
7.3933	0.6579	50	5.3196	0.7748	0.9143	0.8662	0.3864	0.8065	0.9606	0.7587
11.6854	0.7237	55	3.9393	0.8171	0.7311	0.8944	0.3182	0.8441	0.9021	0.4962
7.4125	0.7895	60	3.5938	0.8359	0.7879	0.8662	0.2727	0.8333	0.9068	0.4519
3.9538	0.8553	65	7.5918	0.8246	0.0759	0.8803	0.2955	0.8387	0.5205	0.1650
9.5643	0.9211	70	3.6250	0.8524	0.9310	0.7676	0.1364	0.7903	0.9253	0.5292
6.0378	0.9868	75	4.4797	0.8854	0.9781	0.8662	0.2045	0.8495	0.9781	0.6433
1.8139	1.0526	80	3.2955	0.8656	0.6114	0.8521	0.25	0.8280	0.8418	0.3622
2.0264	1.1184	85	4.3001	0.8007	0.5775	0.7958	0.2955	0.7742	0.7605	0.3922
3.1928	1.1842	90	6.0048	0.8106	0.9730	0.8239	0.2955	0.7957	0.9652	0.8335
4.6055	1.25	95	5.1887	0.8322	0.9495	0.8592	0.2955	0.8226	0.9570	0.7575
0.2403	1.3158	100	3.3475	0.8808	0.5946	0.9085	0.2273	0.8763	0.8870	0.3540
5.5809	1.3816	105	3.6461	0.8880	0.8387	0.8732	0.2045	0.8548	0.9108	0.3595
7.5501	1.4474	110	5.4752	0.8920	0.9991	0.7606	0.0909	0.7957	0.9755	0.5726
0.0361	1.5132	115	4.2694	0.8906	0.7592	0.8028	0.1136	0.8226	0.8370	0.2249
1.0656	1.5789	120	4.6019	0.875	0.9582	0.7817	0.1591	0.7957	0.9066	0.3726
5.2269	1.6447	125	5.1521	0.8552	0.9540	0.8521	0.25	0.8280	0.9289	0.4820
2.4245	1.7105	130	6.7306	0.8497	0.1667	0.8592	0.2727	0.8280	0.7361	0.2013
0.0503	1.7763	135	3.9745	0.8690	0.9344	0.7887	0.1364	0.8065	0.8896	0.3800
8.9398	1.8421	140	4.5412	0.8772	0.9916	0.7817	0.1136	0.8065	0.9578	0.5920
6.6988	1.9079	145	6.2010	0.8580	0.2736	0.8028	0.1818	0.8065	0.6314	0.1502
3.0268	1.9737	150	3.3904	0.8666	0.8661	0.8239	0.1818	0.8226	0.8995	0.4598
8.389	2.0395	155	6.7299	0.8701	0.9924	0.8451	0.2045	0.8333	0.9914	0.8275
2.2611	2.1053	160	5.2049	0.8675	0.9915	0.7535	0.1136	0.7849	0.9769	0.7219
0.713	2.1711	165	3.9889	0.8578	0.7154	0.7535	0.1136	0.7849	0.7864	0.2942
3.0954	2.2368	170	3.5079	0.8828	0.9553	0.7817	0.1136	0.8065	0.9198	0.4591
4.8032	2.3026	175	5.0632	0.8978	0.9977	0.7887	0.0909	0.8172	0.9710	0.5630
4.3986	2.3684	180	4.0567	0.8922	0.9627	0.7887	0.0909	0.8172	0.8991	0.3676
0.4188	2.4342	185	8.6433	0.8816	0.3346	0.7183	0.0682	0.7688	0.6495	0.0959
0.102	2.5	190	4.0822	0.8896	0.9759	0.7042	0.0455	0.7634	0.8706	0.3125
0.0208	2.5658	195	5.6268	0.8921	0.9991	0.7746	0.1136	0.8011	0.9750	0.5927
2.0707	2.6316	200	5.7447	0.8872	0.9965	0.8380	0.1818	0.8333	0.9767	0.5957
7.4128	2.6974	205	4.1685	0.8910	0.9576	0.8310	0.1818	0.8280	0.9228	0.3992
0.4696	2.7632	210	4.6571	0.8903	0.8233	0.7746	0.1364	0.7957	0.8230	0.2367
0.0082	2.8289	215	4.2964	0.8931	0.7931	0.8099	0.1364	0.8226	0.8451	0.2705
3.7742	2.8947	220	4.3192	0.8899	0.8852	0.7535	0.0909	0.7903	0.8429	0.2763
0.0029	2.9605	225	4.8557	0.8905	0.9985	0.7042	0.0682	0.7581	0.9572	0.4598
1.4531	3.0263	230	8.8182	0.8936	0.9999	0.8521	0.1818	0.8441	0.9920	0.7152
5.5001	3.0921	235	8.7725	0.8944	0.9999	0.8239	0.1591	0.8280	0.9864	0.6992
0.0	3.1579	240	8.2296	0.8985	0.9999	0.8451	0.1591	0.8441	0.9808	0.6572
2.672	3.2237	245	6.4457	0.9012	0.9987	0.8662	0.1591	0.8602	0.9663	0.4953
0.001	3.2895	250	5.2402	0.8981	0.9888	0.8310	0.1136	0.8441	0.9245	0.3554
7.0859	3.3553	255	5.5706	0.8969	0.9118	0.8099	0.1136	0.8280	0.8775	0.2516
0.043	3.4211	260	6.0557	0.8968	0.8702	0.8099	0.1136	0.8280	0.8559	0.2258
0.0957	3.4868	265	5.8635	0.8984	0.9951	0.7254	0.0455	0.7796	0.8867	0.2893
0.0549	3.5526	270	6.2190	0.8967	0.9987	0.7324	0.0455	0.7849	0.8956	0.3103
0.0121	3.6184	275	6.3442	0.8925	0.9724	0.8169	0.1364	0.8280	0.8801	0.3023
0.2941	3.6842	280	6.2070	0.8915	0.9612	0.8239	0.1591	0.8280	0.8791	0.3018
4.8077	3.75	285	6.3096	0.8922	0.9813	0.7324	0.0682	0.7796	0.8573	0.2678
0.0033	3.8158	290	6.7836	0.8921	0.9576	0.7183	0.0455	0.7742	0.8302	0.2296
0.8301	3.8816	295	6.3422	0.8962	0.9683	0.7324	0.0455	0.7849	0.8475	0.2449
0.004	3.9474	300	5.7884	0.8962	0.9557	0.8239	0.1591	0.8280	0.8845	0.2867
0.0394	4.0132	305	5.6715	0.8956	0.9972	0.7465	0.0682	0.7903	0.9081	0.3129
0.0002	4.0789	310	5.6708	0.8972	0.9970	0.7465	0.0682	0.7903	0.9118	0.3191
0.1525	4.1447	315	5.7288	0.8971	0.9987	0.7394	0.0682	0.7849	0.9278	0.3364
0.0	4.2105	320	5.8069	0.8944	0.9989	0.7394	0.0682	0.7849	0.9319	0.3477
0.0013	4.2763	325	5.8040	0.8938	0.9894	0.8451	0.1818	0.8387	0.9321	0.3512
0.0118	4.3421	330	5.8521	0.8941	0.9985	0.7606	0.0909	0.7957	0.9353	0.3585
0.0738	4.4079	335	5.8990	0.8918	0.9929	0.8451	0.1818	0.8387	0.9358	0.3606
0.1291	4.4737	340	5.8669	0.8927	0.9902	0.8521	0.1818	0.8441	0.9354	0.3520
0.0072	4.5395	345	5.8455	0.8924	0.9986	0.7394	0.0682	0.7849	0.9308	0.3456
0.0059	4.6053	350	5.8675	0.8923	0.9869	0.8521	0.1818	0.8441	0.9315	0.3486
0.0	4.6711	355	5.8651	0.8937	0.9986	0.7394	0.0682	0.7849	0.9311	0.3437
0.5637	4.7368	360	5.8816	0.8918	0.9838	0.8521	0.1818	0.8441	0.9279	0.3421
0.1122	4.8026	365	5.8021	0.8948	0.9986	0.7394	0.0682	0.7849	0.9271	0.3410
0.0008	4.8684	370	5.8308	0.8941	0.9874	0.8521	0.1818	0.8441	0.9282	0.3385
0.0062	4.9342	375	5.8211	0.8932	0.9843	0.8521	0.1818	0.8441	0.9279	0.3417
0.0017	5.0	380	5.8458	0.8923	0.9836	0.8521	0.1818	0.8441	0.9274	0.3398

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_epo5_lr5e-4_r8_2602041844

bosch_RM_Qwen_S12345_LLM_false_STRUCT_false_epo5_lr5e-4_r8_2602041844

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_epo5_lr5e-4_r8_2602041844