You can deploy this model and start earning money today!

leobianco/
npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo3_lr1e-3_r8_2602031322

Name: npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo3_lr1e-3_r8_2602031322
Availability: InStock

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

apache-2.0

Views

No views yet

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo3_lr1e-3_r8_2602031322

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: 5.3393
Roc Auc: 0.5506
Best Threshold: 0.4844
Tpr At Best Threshold: 0.7949
Fpr At Best Threshold: 0.6
Accuracy At Best Threshold: 0.7143
Avg Score True Positives: 0.5236
Avg Score True Negatives: 0.5125

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: 0.001
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: 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	166.3551	0.4440	0.0000	0.5897	0.5667	0.5578	0.1647	0.1810
138.4375	0.125	5	52.3599	0.4651	0.7186	0.9060	0.9	0.7415	0.9083	0.9364
73.4141	0.25	10	52.7970	0.4336	1.0000	0.3590	0.3	0.4286	0.5963	0.8324
206.0	0.375	15	53.3220	0.5199	1.0000	0.8291	0.7667	0.7075	0.9299	0.9404
191.0	0.5	20	77.9860	0.5667	1.0	1.0	0.8667	0.8231	1.0	0.9990
105.0312	0.625	25	73.3775	0.5071	1.0000	0.9829	0.9667	0.7891	1.0000	1.0000
212.0	0.75	30	103.3878	0.5	inf	0.0	0.0	0.2041	1.0	1.0
308.75	0.875	35	52.9728	0.5	inf	0.0	0.0	0.2041	1.0	1.0
119.6875	1.0	40	222.4014	0.5329	0.0000	0.6239	0.4667	0.6054	0.0000	0.0000
73.6875	1.125	45	98.8503	0.5983	0.0000	0.5556	0.3333	0.5782	0.0000	0.0000
57.2188	1.25	50	8.8449	0.5047	0.9979	0.0940	0.0	0.2789	0.9947	0.9947
22.9094	1.375	55	13.0192	0.6127	0.9996	0.7607	0.5333	0.7007	0.9996	0.9996
47.25	1.5	60	8.4594	0.5823	0.3486	0.2479	0.0333	0.3946	0.2957	0.2717
45.5	1.625	65	8.3806	0.65	0.2451	0.7521	0.4333	0.7143	0.2985	0.2632
19.2266	1.75	70	4.0037	0.5899	0.7432	0.8718	0.5667	0.7823	0.7849	0.7584
8.1635	1.875	75	9.4754	0.4903	0.9946	0.8803	0.7667	0.7483	0.9962	0.9958
8.959	2.0	80	4.1203	0.6283	0.7311	0.4103	0.1667	0.4966	0.7148	0.6929
15.229	2.125	85	7.7125	0.5229	0.9906	0.4957	0.3667	0.5238	0.9900	0.9899
11.9658	2.25	90	17.9511	0.6390	0.0583	0.6325	0.2667	0.6531	0.0607	0.0563
8.2344	2.375	95	5.1694	0.4620	0.6077	0.1709	0.0333	0.3333	0.5491	0.5516
9.7734	2.5	100	6.0364	0.4654	0.4378	0.6239	0.5667	0.5850	0.4583	0.4668
5.749	2.625	105	4.0455	0.5801	0.7982	0.5641	0.3667	0.5782	0.7993	0.7908
5.4023	2.75	110	5.1705	0.5235	0.5234	0.6923	0.6	0.6327	0.5455	0.5402
5.8047	2.875	115	5.2432	0.5439	0.4844	0.8547	0.6667	0.7483	0.5364	0.5258
5.4766	3.0	120	5.3393	0.5506	0.4844	0.7949	0.6	0.7143	0.5236	0.5125

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_epo3_lr1e-3_r8_2602031322

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo3_lr1e-3_r8_2602031322

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_epo3_lr1e-3_r8_2602031322