You can deploy this model and start earning money today!

leobianco/
bosch_RM_mistralai_S130798_LLM_false_STRUCT_true_epo3_lr5e-5_r8_2602051321

Name: bosch_RM_mistralai_S130798_LLM_false_STRUCT_true_epo3_lr5e-5_r8_2602051321
Availability: InStock

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

apache-2.0

Views

No views yet

bosch_RM_mistralai_S130798_LLM_false_STRUCT_true_epo3_lr5e-5_r8_2602051321

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: 24.1273
Roc Auc: 0.7207
Best Threshold: 0.0066
Tpr At Best Threshold: 0.8662
Fpr At Best Threshold: 0.5227
Accuracy At Best Threshold: 0.7742
Avg Score True Positives: 0.7467
Avg Score True Negatives: 0.4541

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: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 130798
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: 3.0

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	79.1626	0.5855	0.0000	0.7887	0.6136	0.6935	0.2545	0.1752
75.9073	0.0820	5	73.1731	0.5891	0.0000	0.7817	0.6136	0.6882	0.2917	0.1993
51.3759	0.1639	10	45.8900	0.6080	0.0000	0.8662	0.6591	0.7419	0.5132	0.3730
59.7658	0.2459	15	29.6685	0.6279	0.9019	0.8380	0.5909	0.7366	0.8709	0.6555
50.2813	0.3279	20	33.6848	0.6400	1.0000	0.8239	0.5227	0.7419	0.9466	0.7609
158.3125	0.4098	25	26.8985	0.6658	0.9879	0.8099	0.5227	0.7312	0.8824	0.6226
61.6094	0.4918	30	39.3861	0.6524	0.0000	0.9225	0.6364	0.7903	0.5160	0.3620
39.9656	0.5738	35	38.0862	0.6657	0.0000	0.8803	0.5682	0.7742	0.5343	0.3415
41.0078	0.6557	40	26.3686	0.6810	0.0018	0.9155	0.5455	0.8065	0.7379	0.4755
8.0507	0.7377	45	24.4861	0.6767	0.0024	0.9155	0.5455	0.8065	0.7587	0.4758
26.375	0.8197	50	25.6766	0.6696	0.0006	0.9085	0.5455	0.8011	0.7245	0.4661
34.5555	0.9016	55	28.9662	0.6733	0.0000	0.8944	0.5455	0.7903	0.6720	0.4304
28.3022	0.9836	60	34.5692	0.6657	0.0000	0.8592	0.5455	0.7634	0.5911	0.3580
18.975	1.0656	65	25.2430	0.6843	0.0010	0.9085	0.5455	0.8011	0.7667	0.4682
16.2547	1.1475	70	24.8862	0.6864	0.0073	0.8873	0.5227	0.7903	0.7617	0.4650
13.6617	1.2295	75	29.5181	0.6933	0.0000	0.8944	0.5227	0.7957	0.6528	0.4131
77.5627	1.3115	80	40.3089	0.6929	0.0000	0.8803	0.5455	0.7796	0.4961	0.2818
0.0363	1.3934	85	27.7714	0.7032	0.0008	0.8380	0.5	0.7581	0.6665	0.4080
14.3438	1.4754	90	23.6903	0.6939	0.9641	0.7746	0.4545	0.7204	0.8223	0.5386
6.8443	1.5574	95	28.0154	0.6935	0.9848	0.6197	0.3182	0.6344	0.6991	0.4537
12.8916	1.6393	100	42.8714	0.6749	0.0061	0.6408	0.3636	0.6398	0.5020	0.2845
3.3044	1.7213	105	40.6510	0.6795	0.0087	0.6479	0.3636	0.6452	0.5243	0.3036
11.7398	1.8033	110	26.9135	0.6981	0.9659	0.6408	0.3182	0.6505	0.7320	0.4775
8.4022	1.8852	115	24.0780	0.7039	1.0000	0.5775	0.2273	0.6237	0.8571	0.5669
7.4942	1.9672	120	24.4091	0.7169	1.0000	0.6408	0.2727	0.6613	0.8943	0.6185
3.9178	2.0492	125	23.6358	0.7119	0.9372	0.8521	0.5	0.7688	0.8712	0.5665
0.0018	2.1311	130	23.3927	0.7129	0.4961	0.8662	0.5227	0.7742	0.8492	0.5236
15.0474	2.2131	135	23.9107	0.7140	0.2846	0.8239	0.4773	0.7527	0.8002	0.4840
1.2848	2.2951	140	23.8801	0.7158	0.0140	0.8662	0.5227	0.7742	0.7697	0.4715
0.0109	2.3770	145	24.5023	0.7195	0.9975	0.6268	0.2727	0.6505	0.7521	0.4609
0.0	2.4590	150	24.5563	0.7191	0.9965	0.6268	0.2727	0.6505	0.7475	0.4564
15.5717	2.5410	155	24.7714	0.7170	0.9912	0.6338	0.2727	0.6559	0.7415	0.4519
0.9453	2.6230	160	24.8286	0.7162	0.9938	0.6127	0.2727	0.6398	0.7372	0.4499
0.057	2.7049	165	24.9657	0.7144	0.0081	0.8451	0.5	0.7634	0.7366	0.4551
0.0643	2.7869	170	24.7759	0.7170	0.0051	0.8662	0.5227	0.7742	0.7407	0.4538
35.7505	2.8689	175	24.5529	0.7169	0.9950	0.6268	0.2727	0.6505	0.7432	0.4578
0.4749	2.9508	180	24.1273	0.7207	0.0066	0.8662	0.5227	0.7742	0.7467	0.4541

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_S130798_LLM_false_STRUCT_true_epo3_lr5e-5_r8_2602051321

bosch_RM_mistralai_S130798_LLM_false_STRUCT_true_epo3_lr5e-5_r8_2602051321

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

leobianco/
bosch_RM_mistralai_S130798_LLM_false_STRUCT_true_epo3_lr5e-5_r8_2602051321