You can deploy this model and start earning money today!

leobianco/
bosch_RM_mistralai_S130104_LLM_false_STRUCT_false_epo5_lr2e-4_r16_2602061047

Name: bosch_RM_mistralai_S130104_LLM_false_STRUCT_false_epo5_lr2e-4_r16_2602061047
Availability: InStock

mistralai/Mistral-7B-Instruct-v0.3

apache-2.0

Views

No views yet

bosch_RM_mistralai_S130104_LLM_false_STRUCT_false_epo5_lr2e-4_r16_2602061047

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 130.3277
Roc Auc: 0.8145
Best Threshold: 0.0000
Tpr At Best Threshold: 0.9648
Fpr At Best Threshold: 0.3409
Accuracy At Best Threshold: 0.8925
Avg Score True Positives: 0.9344
Avg Score True Negatives: 0.3343

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: 0.0002
train_batch_size: 1
eval_batch_size: 1
seed: 130104
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	788.1682	0.5639	0.0000	0.4859	0.3409	0.5269	0.2481	0.1817
469.0	0.0658	5	484.3256	0.5866	0.0000	0.7183	0.5455	0.6559	0.4850	0.3420
1571.5	0.1316	10	547.2258	0.5114	1.0	1.0	0.9773	0.7688	1.0	0.9773
240.0	0.1974	15	256.5260	0.6777	0.9999	0.9366	0.5682	0.8172	0.9424	0.6099
760.0	0.2632	20	266.2209	0.7025	0.0000	0.8803	0.5	0.7903	0.6411	0.3184
499.0	0.3289	25	268.8077	0.6192	0.9984	0.9507	0.7045	0.7957	0.9574	0.7458
149.0	0.3947	30	198.6879	0.7267	0.9978	0.8803	0.4318	0.8065	0.8805	0.4407
283.5	0.4605	35	422.9301	0.5807	0.6876	0.9859	0.8182	0.7957	0.9837	0.8182
221.0625	0.5263	40	280.6357	0.6612	0.0000	0.8380	0.5227	0.7527	0.8242	0.5227
385.0	0.5921	45	180.0667	0.7244	1.0000	0.8521	0.4318	0.7849	0.8663	0.4773
1343.0	0.6579	50	343.7581	0.6332	1.0	0.9930	0.7273	0.8226	0.9930	0.7273
0.0	0.7237	55	189.0283	0.7634	1.0000	0.9155	0.4091	0.8387	0.9279	0.4552
454.75	0.7895	60	407.8714	0.7186	0.0000	0.7465	0.3182	0.7312	0.5599	0.2285
0.0	0.8553	65	274.7845	0.6429	1.0	0.9648	0.6818	0.8118	0.9661	0.7299
6.0	0.9211	70	208.3859	0.6664	0.0000	0.9507	0.5909	0.8226	0.9296	0.5909
667.0	0.9868	75	316.4060	0.6185	0.9992	0.9718	0.7273	0.8065	0.9718	0.7499
215.0	1.0526	80	299.0363	0.6178	1.0000	0.9437	0.7045	0.7903	0.9437	0.7276
0.0	1.1184	85	229.8901	0.7244	1.0000	0.9014	0.4545	0.8172	0.9155	0.5218
200.0	1.1842	90	241.4870	0.7746	1.0000	0.7676	0.2727	0.7581	0.7903	0.3238
619.75	1.25	95	375.1736	0.7673	0.0001	0.6972	0.1818	0.7258	0.6774	0.1818
0.0	1.3158	100	642.6989	0.5795	1.0	1.0	0.8409	0.8011	1.0	0.8409
876.0	1.3816	105	265.6818	0.7588	0.9627	0.9155	0.4091	0.8387	0.9153	0.4136
219.3203	1.4474	110	245.4019	0.8003	0.0000	0.8451	0.2727	0.8172	0.7817	0.25
0.0	1.5132	115	284.3775	0.7366	1.0	0.9930	0.5227	0.8710	0.9930	0.5303
0.0	1.5789	120	165.9576	0.8322	0.9996	0.8944	0.25	0.8602	0.8944	0.2503
60.0	1.6447	125	171.5054	0.7890	0.9998	0.9437	0.3636	0.8710	0.9437	0.3636
0.0	1.7105	130	282.6562	0.6663	0.0000	0.9648	0.6364	0.8226	0.9460	0.6364
73.0693	1.7763	135	186.7155	0.7516	1.0	0.8451	0.3636	0.7957	0.8732	0.4318
0.0016	1.8421	140	253.8056	0.6523	0.9996	0.9648	0.6591	0.8172	0.9648	0.6633
286.5	1.9079	145	242.5628	0.7940	0.0000	0.7254	0.1818	0.7473	0.6366	0.1364
0.0	1.9737	150	454.3232	0.5682	1.0	1.0	0.8636	0.7957	1.0	0.9312
0.0	2.0395	155	264.3398	0.8279	0.0000	0.8310	0.2045	0.8226	0.6262	0.1136
133.5	2.1053	160	134.7479	0.8148	1.0	0.8944	0.2727	0.8548	0.9084	0.3422
404.0	2.1711	165	233.9113	0.6418	1.0	0.9859	0.7045	0.8226	0.9920	0.7470
109.5	2.2368	170	203.3215	0.8500	0.0000	0.8380	0.2045	0.8280	0.6823	0.1136
0.0	2.3026	175	159.9825	0.8379	0.0000	0.8944	0.25	0.8602	0.8310	0.2273
22.625	2.3684	180	296.7601	0.8318	0.0000	0.7887	0.1818	0.7957	0.6858	0.1136
86.9375	2.4342	185	176.2527	0.8075	0.0000	0.8944	0.2955	0.8495	0.8662	0.2727
1.625	2.5	190	349.3645	0.8047	0.0000	0.7606	0.2045	0.7688	0.5826	0.1054
153.0	2.5658	195	124.4285	0.8236	0.0000	0.9366	0.3409	0.8710	0.8469	0.2732
328.0	2.6316	200	236.4281	0.6560	1.0	0.9930	0.6818	0.8333	0.9930	0.7191
153.0	2.6974	205	164.6813	0.7351	1.0000	0.9648	0.5	0.8548	0.9807	0.5366
0.0	2.7632	210	144.0057	0.8183	0.0000	0.9296	0.3409	0.8656	0.8377	0.2702
0.0	2.8289	215	191.1938	0.8220	0.0000	0.8662	0.2727	0.8333	0.7360	0.2273
0.0	2.8947	220	181.0903	0.7426	1.0	0.9789	0.5	0.8656	0.9848	0.5424
117.5	2.9605	225	230.4652	0.7159	1.0	1.0	0.5682	0.8656	1.0	0.6148
0.0	3.0263	230	137.3762	0.7898	0.9980	0.9085	0.3636	0.8441	0.9103	0.3863
0.0	3.0921	235	147.7886	0.8062	1.0	0.8310	0.2727	0.8065	0.8541	0.3182
0.0	3.1579	240	138.7992	0.7958	1.0	0.8944	0.3409	0.8387	0.9150	0.3639
0.0	3.2237	245	150.8022	0.7732	1.0	0.9296	0.4091	0.8495	0.9435	0.4318
324.5	3.2895	250	162.4785	0.7642	0.9996	0.9507	0.4318	0.8602	0.9507	0.4318
102.0	3.3553	255	143.0079	0.8108	0.0000	0.9225	0.3182	0.8656	0.9010	0.3182
0.0	3.4211	260	149.6112	0.8251	0.0000	0.8732	0.25	0.8441	0.8394	0.25
103.5	3.4868	265	145.7514	0.8259	0.0000	0.8873	0.25	0.8548	0.8459	0.25
0.0	3.5526	270	138.9257	0.8327	1.0000	0.8873	0.25	0.8548	0.8907	0.2724
0.0	3.6184	275	144.0393	0.8048	0.0000	0.9577	0.3636	0.8817	0.9367	0.3636
0.0	3.6842	280	139.0020	0.8331	1.0	0.9085	0.2727	0.8656	0.9085	0.3377
592.0	3.75	285	139.4355	0.8276	0.9998	0.9085	0.2727	0.8656	0.9084	0.2727
0.0	3.8158	290	140.4583	0.8372	1.0	0.8873	0.25	0.8548	0.8966	0.2727
0.0	3.8816	295	136.4138	0.8326	1.0	0.9014	0.2727	0.8602	0.9084	0.3185
0.0	3.9474	300	135.1710	0.8348	1.0	0.9014	0.2727	0.8602	0.9176	0.3266
0.0	4.0132	305	133.7296	0.8253	0.0000	0.9577	0.3409	0.8871	0.9173	0.3409
48.0	4.0789	310	133.2589	0.8211	0.0000	0.9577	0.3409	0.8871	0.9203	0.3409
0.0	4.1447	315	135.5546	0.8153	0.0000	0.9577	0.3409	0.8871	0.9419	0.3409
0.0	4.2105	320	136.0417	0.8164	0.0213	0.9577	0.3409	0.8871	0.9508	0.3409
0.0	4.2763	325	136.7820	0.8122	0.6514	0.9648	0.3409	0.8925	0.9623	0.3409
0.0	4.3421	330	132.9046	0.8146	0.0002	0.9648	0.3409	0.8925	0.9505	0.3409
0.0	4.4079	335	130.0684	0.8139	0.0000	0.9648	0.3409	0.8925	0.9413	0.3405
0.0	4.4737	340	129.3722	0.8119	0.0000	0.9648	0.3409	0.8925	0.9162	0.3210
0.0	4.5395	345	128.9759	0.8139	0.0000	0.9648	0.3409	0.8925	0.9235	0.3353
0.0	4.6053	350	129.5495	0.8107	0.0141	0.9437	0.3182	0.8817	0.9203	0.3185
0.0	4.6711	355	129.7174	0.8139	0.0000	0.9648	0.3409	0.8925	0.9362	0.3374
95.5312	4.7368	360	130.4631	0.8145	0.9047	0.9437	0.3182	0.8817	0.9428	0.3384
0.0	4.8026	365	130.5971	0.8143	0.0000	0.9648	0.3409	0.8925	0.9335	0.3206
0.0	4.8684	370	129.3719	0.8245	0.1480	0.9437	0.3182	0.8817	0.9352	0.3186
0.0	4.9342	375	129.0382	0.8143	0.3775	0.9437	0.3182	0.8817	0.9351	0.3214
0.0	5.0	380	130.3277	0.8145	0.0000	0.9648	0.3409	0.8925	0.9344	0.3343

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_mistralai_S130104_LLM_false_STRUCT_false_epo5_lr2e-4_r16_2602061047

bosch_RM_mistralai_S130104_LLM_false_STRUCT_false_epo5_lr2e-4_r16_2602061047

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

leobianco/
bosch_RM_mistralai_S130104_LLM_false_STRUCT_false_epo5_lr2e-4_r16_2602061047