You can deploy this model and start earning money today!

leobianco/
npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602251336

Name: npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602251336
Availability: InStock

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

apache-2.0

Views

No views yet

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602251336

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: 11.3037
Roc Auc: 0.8771
Best Threshold: 1.0000
Tpr At Best Threshold: 0.8120
Fpr At Best Threshold: 0.1333
Accuracy At Best Threshold: 0.8231
Avg Score True Positives: 0.9506
Avg Score True Negatives: 0.4710

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: 0.0001
train_batch_size: 2
eval_batch_size: 2
seed: 12345
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 16
total_eval_batch_size: 16
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	161.0884	0.6632	0.0000	0.5043	0.1667	0.5714	0.0112	0.0000
86.1875	0.125	5	76.1479	0.6566	0.0000	0.5043	0.1667	0.5714	0.1824	0.0011
64.0938	0.25	10	36.3912	0.4821	0.9952	1.0	0.9667	0.8027	0.9999	0.9998
24.424	0.375	15	32.4290	0.6593	0.0172	0.5128	0.1667	0.5782	0.2596	0.0618
18.2001	0.5	20	11.2938	0.6231	0.9974	0.4786	0.1667	0.5510	0.8294	0.8377
8.207	0.625	25	6.7254	0.6799	0.8418	0.4444	0.0333	0.5510	0.6501	0.4952
10.1719	0.75	30	4.9399	0.7098	0.6507	0.4701	0.0667	0.5646	0.5934	0.4556
6.4258	0.875	35	6.5778	0.7443	0.3630	0.6410	0.2	0.6735	0.4080	0.3028
5.7773	1.0	40	5.5389	0.7943	0.3775	0.8291	0.3667	0.7891	0.4825	0.3674
5.6875	1.125	45	4.2239	0.7523	0.6320	0.5128	0.0667	0.5986	0.6391	0.5365
5.1914	1.25	50	4.2865	0.7477	0.6114	0.5128	0.1	0.5918	0.6285	0.5258
4.2227	1.375	55	5.2445	0.7919	0.5146	0.4359	0.0	0.5510	0.5101	0.3899
4.7031	1.5	60	4.4739	0.8459	0.5487	0.5983	0.0667	0.6667	0.5801	0.4083
5.5566	1.625	65	7.4192	0.8611	0.2422	0.8462	0.2667	0.8231	0.3471	0.1896
7.5059	1.75	70	2.8692	0.8741	0.8710	0.7350	0.1	0.7687	0.8848	0.6306
5.0684	1.875	75	3.6072	0.8466	0.5983	0.7607	0.1667	0.7755	0.6883	0.3592
6.9348	2.0	80	2.8397	0.8440	0.9252	0.6410	0.1	0.6939	0.8931	0.5571
2.377	2.125	85	2.8963	0.8607	0.8278	0.6667	0.0667	0.7211	0.8112	0.4156
2.0393	2.25	90	3.9185	0.8786	0.6289	0.7179	0.0667	0.7619	0.7150	0.2443
0.5402	2.375	95	2.6007	0.9179	0.9155	0.8120	0.0667	0.8367	0.9112	0.4137
3.0466	2.5	100	4.2253	0.9322	0.9974	0.8120	0.0667	0.8367	0.9627	0.5705
1.7867	2.625	105	3.9971	0.9131	0.2862	0.9231	0.2	0.8980	0.7972	0.1910
1.6476	2.75	110	4.5581	0.9164	0.9684	0.9060	0.2	0.8844	0.9478	0.4682
3.2442	2.875	115	5.6066	0.9199	0.9956	0.9145	0.1667	0.8980	0.9604	0.4959
0.0072	3.0	120	6.1045	0.9158	0.9972	0.9060	0.1667	0.8912	0.9575	0.4728
2.0881	3.125	125	6.1382	0.9120	0.9949	0.8803	0.1667	0.8707	0.9405	0.4095
0.5253	3.25	130	7.0466	0.9128	0.9996	0.8547	0.1667	0.8503	0.9594	0.4511
0.0258	3.375	135	6.8565	0.8979	0.9904	0.8034	0.1667	0.8095	0.9083	0.3425
0.0	3.5	140	11.3937	0.8813	1.0000	0.8889	0.2333	0.8639	0.9810	0.5127
7.1094	3.625	145	12.9939	0.8460	1.0000	0.8718	0.2333	0.8503	0.9848	0.5229
0.5626	3.75	150	9.1369	0.9001	1.0000	0.8205	0.1667	0.8231	0.9369	0.4108
0.0005	3.875	155	9.5878	0.8954	0.9996	0.7692	0.1333	0.7891	0.9116	0.3523
0.0741	4.0	160	9.9258	0.8917	0.9998	0.7863	0.1333	0.8027	0.9208	0.3672
0.0	4.125	165	10.2300	0.8977	1.0000	0.7863	0.1333	0.8027	0.9303	0.3911
0.0	4.25	170	10.4572	0.8826	1.0000	0.8291	0.1667	0.8299	0.9419	0.4439
0.0001	4.375	175	11.1382	0.8762	1.0000	0.8632	0.1667	0.8571	0.9515	0.4723
0.0	4.5	180	11.3789	0.8805	1.0000	0.8462	0.1667	0.8435	0.9567	0.4792
0.0004	4.625	185	11.3422	0.8779	1.0000	0.8462	0.1667	0.8435	0.9531	0.4750
0.0021	4.75	190	11.3606	0.8775	1.0000	0.8462	0.1667	0.8435	0.9525	0.4773
0.0	4.875	195	11.3102	0.8762	1.0000	0.8462	0.1667	0.8435	0.9515	0.4664
0.0001	5.0	200	11.3037	0.8771	1.0000	0.8120	0.1333	0.8231	0.9506	0.4710

Framework versions

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

leobianco/npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602251336

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602251336

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

leobianco/
npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602251336