You can deploy this model and start earning money today!

leobianco/
npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602271027

Name: npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602271027
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_2602271027

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: 12.5002
Roc Auc: 0.8647
Best Threshold: 1.0000
Tpr At Best Threshold: 0.9316
Fpr At Best Threshold: 0.2333
Accuracy At Best Threshold: 0.8980
Avg Score True Positives: 0.9643
Avg Score True Negatives: 0.3872

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	135.9014	0.6534	0.0000	0.6325	0.3	0.6463	0.0195	0.0000
71.4688	0.125	5	54.2886	0.6652	0.0012	0.4957	0.1	0.5782	0.2036	0.0340
66.5625	0.25	10	38.2962	0.4765	0.9999	0.9487	0.9333	0.7687	0.9922	1.0000
23.8259	0.375	15	32.0047	0.6523	0.0083	0.6239	0.3	0.6395	0.2610	0.0548
27.0182	0.5	20	12.3850	0.6165	0.9994	0.6154	0.3	0.6327	0.9000	0.9143
13.5469	0.625	25	8.9547	0.6618	0.4254	0.5641	0.2	0.6122	0.4326	0.2911
11.3887	0.75	30	9.3559	0.6147	0.3639	0.5043	0.2	0.5646	0.3327	0.2645
7.1396	0.875	35	4.0949	0.6107	0.8146	0.5385	0.3333	0.5646	0.7709	0.7011
5.3574	1.0	40	4.0207	0.6825	0.7058	0.5556	0.2667	0.5918	0.6987	0.5910
9.793	1.125	45	3.5166	0.7764	0.7807	0.5726	0.1333	0.6327	0.7748	0.6416
4.8633	1.25	50	3.7249	0.8128	0.6834	0.6667	0.1333	0.7075	0.6990	0.5448
4.3516	1.375	55	4.5569	0.8285	0.5254	0.7094	0.1667	0.7347	0.5689	0.3865
4.334	1.5	60	5.6224	0.8530	0.3803	0.7778	0.1333	0.7959	0.4672	0.2608
4.8516	1.625	65	4.3035	0.8336	0.4455	0.8632	0.3	0.8299	0.6001	0.3273
4.0254	1.75	70	4.3828	0.8118	0.5059	0.7607	0.2333	0.7619	0.6265	0.2997
2.9897	1.875	75	2.7805	0.8803	0.9284	0.8034	0.1333	0.8163	0.9333	0.5422
2.7801	2.0	80	10.3882	0.8840	0.0336	0.9145	0.2667	0.8776	0.3319	0.0498
2.3579	2.125	85	5.9515	0.9132	0.9993	0.8462	0.1333	0.8503	0.9858	0.7016
1.0946	2.25	90	5.9650	0.8822	0.3451	0.7778	0.1333	0.7959	0.6781	0.1530
0.2981	2.375	95	3.6888	0.9238	0.9433	0.9316	0.2	0.9048	0.9496	0.4202
1.0804	2.5	100	8.4156	0.9472	1.0000	0.8632	0.1	0.8707	0.9992	0.6418
5.2071	2.625	105	4.6044	0.9417	0.9985	0.8803	0.1333	0.8776	0.9835	0.5500
0.4578	2.75	110	2.5156	0.9308	0.7853	0.9402	0.2333	0.9048	0.9401	0.3737
0.4978	2.875	115	2.6707	0.9279	0.8429	0.8632	0.1333	0.8639	0.9027	0.2711
0.0101	3.0	120	4.5611	0.9309	0.9972	0.9145	0.2	0.8912	0.9750	0.5117
0.8758	3.125	125	5.6226	0.9291	0.9989	0.9231	0.2	0.8980	0.9694	0.4626
0.0123	3.25	130	7.7723	0.9154	1.0000	0.9402	0.2	0.9116	0.9719	0.4498
0.0	3.375	135	8.8573	0.9024	0.9999	0.9487	0.2333	0.9116	0.9638	0.3990
0.0	3.5	140	9.7488	0.8996	0.9996	0.9402	0.2333	0.9048	0.9589	0.3757
0.0116	3.625	145	11.0751	0.8657	1.0000	0.9316	0.2333	0.8980	0.9681	0.3968
0.0163	3.75	150	10.7528	0.8880	1.0000	0.9316	0.2333	0.8980	0.9605	0.3740
0.0	3.875	155	11.8625	0.8607	1.0000	0.9316	0.2333	0.8980	0.9591	0.3686
3.9376	4.0	160	11.7211	0.8621	1.0000	0.9316	0.2333	0.8980	0.9653	0.3789
0.0	4.125	165	12.0375	0.8623	1.0000	0.9231	0.2333	0.8912	0.9643	0.3761
0.0	4.25	170	11.7372	0.8615	1.0000	0.9316	0.2333	0.8980	0.9635	0.3748
0.0008	4.375	175	11.8182	0.8608	1.0000	0.9316	0.2333	0.8980	0.9625	0.3743
0.0	4.5	180	11.9840	0.8625	1.0000	0.9316	0.2333	0.8980	0.9648	0.3773
0.0	4.625	185	12.5588	0.8496	1.0000	0.9316	0.2333	0.8980	0.9649	0.3868
0.0	4.75	190	12.4359	0.8640	1.0000	0.9231	0.2333	0.8912	0.9651	0.3882
0.0001	4.875	195	12.7064	0.8644	1.0000	0.9145	0.2333	0.8844	0.9650	0.3895
0.0	5.0	200	12.5002	0.8647	1.0000	0.9316	0.2333	0.8980	0.9643	0.3872

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_2602271027

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602271027

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_2602271027