learning_rate: 0.001
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
25.6526	0.0658	5	26.8026	0.6140	0.1366	0.4507	0.1818	0.5376	0.3007	0.1080
14.4531	0.1316	10	16.4415	0.6378	1.0000	0.4718	0.1364	0.5645	0.9486	0.9464
0.0004	0.1974	15	27.2906	0.6266	1.0000	0.5775	0.3409	0.5968	0.9999	0.9784
20.6881	0.2632	20	9.7832	0.6438	0.6926	0.4014	0.1364	0.5108	0.5002	0.3209
20.3781	0.3289	25	7.2734	0.6824	0.9938	0.5	0.1364	0.5860	0.9250	0.8578
5.5999	0.3947	30	16.8174	0.7654	0.0216	0.7042	0.25	0.7151	0.3339	0.1284
19.8977	0.4605	35	5.9490	0.7757	0.9798	0.6338	0.1364	0.6882	0.8712	0.5771
8.8756	0.5263	40	4.3199	0.8400	0.8832	0.8662	0.2727	0.8333	0.9244	0.4720
12.917	0.5921	45	5.1573	0.8234	0.6775	0.8310	0.2727	0.8065	0.8444	0.3451
12.0892	0.6579	50	12.1926	0.5106	0.9997	0.3380	0.1818	0.4516	0.9398	0.9491
11.9426	0.7237	55	4.5976	0.5252	0.8392	0.2958	0.1818	0.4194	0.7670	0.7626
12.7847	0.7895	60	7.8125	0.4437	0.9987	0.0070	0.0	0.2419	0.9623	0.9746
8.9551	0.8553	65	7.4602	0.5209	0.9852	0.3732	0.2273	0.4677	0.9723	0.9719
11.0557	0.9211	70	4.8452	0.5932	0.9096	0.3592	0.1136	0.4839	0.7749	0.7218
22.8126	0.9868	75	23.4759	0.4226	1.0000	0.9789	0.9091	0.7688	1.0000	1.0000
5.8896	1.0526	80	8.3955	0.5045	0.9912	0.4155	0.2727	0.4892	0.9574	0.9748
7.5889	1.1184	85	15.3642	0.3994	0.2790	0.9859	0.9318	0.7688	0.9770	0.9358
11.6089	1.1842	90	7.0293	0.4144	0.9936	0.0634	0.0455	0.2742	0.8396	0.8895
9.9492	1.25	95	5.5723	0.4529	0.8693	0.2746	0.2045	0.3978	0.6848	0.7266
3.8242	1.3158	100	7.9450	0.4525	0.9930	0.0493	0.0	0.2742	0.9795	0.9819
12.9546	1.3816	105	4.8435	0.4251	0.7186	0.9789	0.9318	0.7634	0.8412	0.8535
5.6992	1.4474	110	13.6841	0.5418	0.1128	0.5	0.3864	0.5269	0.1157	0.1101
8.0012	1.5132	115	9.0155	0.4171	0.9943	0.0704	0.0227	0.2849	0.9900	0.9907
2.9248	1.5789	120	6.1812	0.4483	0.4922	0.3592	0.3409	0.4301	0.4495	0.4625
10.2742	1.6447	125	6.4614	0.4980	0.7773	0.1620	0.0455	0.3495	0.4871	0.4782
4.7529	1.7105	130	5.3982	0.4977	0.6654	0.1127	0.0	0.3226	0.5373	0.5379
0.0187	1.7763	135	17.2131	0.5079	0.9999	0.4155	0.2955	0.4839	0.9999	0.9999
10.5627	1.8421	140	15.8160	0.5706	0.9998	0.6408	0.4091	0.6290	0.9998	0.9998
20.1722	1.9079	145	18.8218	0.4546	0.0353	0.8732	0.8182	0.7097	0.0490	0.0508
4.2188	1.9737	150	6.3369	0.4506	0.3551	0.9366	0.8864	0.7419	0.4248	0.4296
9.2405	2.0395	155	6.8023	0.4161	0.9239	1.0	0.9773	0.7688	0.9632	0.9664
6.4674	2.1053	160	4.6654	0.5194	0.7896	0.9789	0.8409	0.7849	0.8255	0.8246
15.8982	2.1711	165	5.4473	0.6376	0.8559	0.7042	0.4318	0.6720	0.8971	0.8500
6.7227	2.2368	170	6.7549	0.6476	0.9659	0.7817	0.5	0.7151	0.9719	0.9654
2.9526	2.3026	175	4.3560	0.6230	0.7786	0.7042	0.4318	0.6720	0.7868	0.7818
3.6641	2.3684	180	4.4168	0.5070	0.7215	0.4789	0.3864	0.5108	0.7192	0.7188
3.0176	2.4342	185	5.3028	0.4514	0.9281	0.0563	0.0	0.2796	0.9200	0.9205
6.7954	2.5	190	4.7308	0.5218	0.8804	0.2254	0.0682	0.3925	0.8741	0.8728
0.9312	2.5658	195	4.9278	0.4935	0.8973	0.4366	0.3182	0.4946	0.8932	0.8928
6.6655	2.6316	200	4.5135	0.4705	0.8685	0.1056	0.0	0.3172	0.8231	0.8251
4.7593	2.6974	205	4.4864	0.4891	0.8584	0.1761	0.0909	0.3495	0.7791	0.7796
4.1416	2.7632	210	4.4454	0.5082	0.6762	0.8310	0.7045	0.7043	0.7200	0.7163
4.5312	2.8289	215	4.7350	0.4999	0.9340	0.1268	0.0455	0.3226	0.8470	0.8464
6.5605	2.8947	220	6.3295	0.5376	0.3567	0.8380	0.6818	0.7151	0.4290	0.4202
0.7766	2.9605	225	7.0538	0.5521	0.9648	0.8380	0.6364	0.7258	0.9738	0.9718
4.6459	3.0263	230	9.1129	0.5459	0.9897	0.8592	0.6818	0.7312	0.9919	0.9913
4.6857	3.0921	235	8.5049	0.5457	0.9866	0.8592	0.7045	0.7258	0.9887	0.9882
0.6826	3.1579	240	4.7424	0.5345	0.8625	0.9085	0.7045	0.7634	0.8767	0.8743
3.3779	3.2237	245	4.3989	0.5288	0.7064	0.7324	0.6136	0.6505	0.7197	0.7157
1.7568	3.2895	250	4.5371	0.4999	0.8179	0.9296	0.8409	0.7473	0.8412	0.8401
10.5361	3.3553	255	4.5996	0.5533	0.8652	0.4155	0.2955	0.4839	0.8591	0.8550
3.0088	3.4211	260	4.8951	0.5855	0.9016	0.4930	0.3409	0.5323	0.8982	0.8905
3.4761	3.4868	265	4.4462	0.5991	0.8087	0.8169	0.6364	0.7097	0.8444	0.8277
2.3599	3.5526	270	4.6195	0.5997	0.8417	0.8239	0.6364	0.7151	0.8757	0.8556
2.437	3.6184	275	4.2755	0.6201	0.7629	0.7465	0.5227	0.6828	0.8084	0.7703
4.7168	3.6842	280	4.3808	0.6148	0.6197	0.7394	0.5227	0.6774	0.7105	0.6576
4.4741	3.75	285	4.6663	0.6052	0.5494	0.6761	0.5	0.6344	0.6493	0.5924
1.8899	3.8158	290	4.3675	0.6144	0.7515	0.5563	0.3636	0.5753	0.7747	0.7226
5.9375	3.8816	295	4.4181	0.6297	0.7684	0.6127	0.4091	0.6075	0.8093	0.7487
7.4058	3.9474	300	5.4562	0.6328	0.9067	0.6690	0.4318	0.6452	0.9215	0.8827
4.2471	4.0132	305	4.5366	0.6293	0.8184	0.6056	0.3864	0.6075	0.8326	0.7723
2.1204	4.0789	310	4.4405	0.6375	0.6348	0.6479	0.3864	0.6398	0.7128	0.6301
4.1409	4.1447	315	4.5666	0.6455	0.8196	0.6761	0.3864	0.6613	0.8560	0.7940
1.728	4.2105	320	4.7445	0.6555	0.8604	0.6690	0.3864	0.6559	0.8858	0.8284
1.9644	4.2763	325	4.4814	0.6557	0.8332	0.6620	0.3864	0.6505	0.8545	0.7846
5.2936	4.3421	330	4.3992	0.6520	0.8029	0.6620	0.3864	0.6505	0.8342	0.7580
2.8845	4.4079	335	4.2855	0.6519	0.7718	0.6197	0.3636	0.6237	0.7877	0.7041
4.0088	4.4737	340	4.3099	0.6513	0.7021	0.6549	0.3864	0.6452	0.7423	0.6493
2.7476	4.5395	345	4.3012	0.6519	0.7038	0.6549	0.3864	0.6452	0.7473	0.6541
3.5598	4.6053	350	4.2700	0.6529	0.7443	0.6549	0.3864	0.6452	0.7765	0.6844
2.4282	4.6711	355	4.2795	0.6560	0.7311	0.7042	0.4545	0.6667	0.7997	0.7108
4.8151	4.7368	360	4.3262	0.6527	0.7364	0.7254	0.4773	0.6774	0.8150	0.7303
1.7397	4.8026	365	4.3250	0.6547	0.7873	0.6549	0.3864	0.6452	0.8122	0.7270
1.7361	4.8684	370	4.3505	0.6541	0.7739	0.6761	0.4318	0.6505	0.8201	0.7352
4.7466	4.9342	375	4.3660	0.6539	0.8081	0.6479	0.3864	0.6398	0.8240	0.7401
3.6379	5.0	380	4.3630	0.6545	0.8038	0.6549	0.3864	0.6452	0.8245	0.7402

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_lr1e-3_r8_2602051203

bosch_RM_Qwen_S12345_LLM_false_STRUCT_false_epo5_lr1e-3_r8_2602051203

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_lr1e-3_r8_2602051203