You can deploy this model and start earning money today!

leobianco/
npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602261705

Name: npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602261705
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_2602261705

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: 9.7280
Roc Auc: 0.8987
Best Threshold: 1.0000
Tpr At Best Threshold: 0.8974
Fpr At Best Threshold: 0.1667
Accuracy At Best Threshold: 0.8844
Avg Score True Positives: 0.9659
Avg Score True Negatives: 0.3777

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	159.7755	0.6771	0.0000	0.5043	0.1667	0.5714	0.0112	0.0000
80.25	0.125	5	73.7134	0.6717	0.0000	0.5214	0.1667	0.5850	0.1574	0.0114
62.7812	0.25	10	37.9116	0.4778	1.0000	0.9402	0.9333	0.7619	0.9960	1.0000
26.906	0.375	15	29.2584	0.6670	0.0478	0.4957	0.1333	0.5714	0.3375	0.0769
17.7128	0.5	20	9.9355	0.6292	0.9980	0.3761	0.1	0.4830	0.8464	0.8445
8.8057	0.625	25	5.2492	0.6691	0.9511	0.5299	0.1667	0.5918	0.8444	0.7840
12.9395	0.75	30	12.7573	0.6362	0.1343	0.6239	0.3	0.6395	0.1730	0.1272
6.6719	0.875	35	4.4055	0.5934	0.6151	0.6923	0.5333	0.6463	0.6609	0.6035
5.2422	1.0	40	4.7412	0.6302	0.5078	0.8205	0.6	0.7347	0.5868	0.5240
5.6094	1.125	45	5.2392	0.6812	0.4571	0.7521	0.4333	0.7143	0.5255	0.4479
5.2969	1.25	50	4.9067	0.7063	0.5059	0.6752	0.3333	0.6735	0.5571	0.4596
5.582	1.375	55	5.6544	0.7744	0.4187	0.7350	0.2667	0.7347	0.4741	0.3379
5.0547	1.5	60	4.5408	0.8219	0.4805	0.8376	0.3	0.8095	0.5732	0.3799
4.5039	1.625	65	5.2724	0.8308	0.3585	0.8803	0.3333	0.8367	0.5054	0.2792
3.2388	1.75	70	4.7707	0.8311	0.4282	0.8034	0.2667	0.7891	0.5679	0.2709
2.54	1.875	75	2.5680	0.8973	0.7217	0.8291	0.1667	0.8299	0.8310	0.3911
1.9219	2.0	80	5.1822	0.9019	0.3821	0.7778	0.1	0.8027	0.6100	0.1344
1.2063	2.125	85	3.5709	0.9205	0.9898	0.8889	0.1667	0.8776	0.9632	0.5383
0.1589	2.25	90	5.7430	0.9093	0.1755	0.9060	0.2	0.8844	0.7578	0.1457
0.0181	2.375	95	5.2869	0.9302	0.9720	0.9316	0.1667	0.9116	0.9482	0.3801
4.5628	2.5	100	10.2819	0.9368	1.0000	0.9316	0.1667	0.9116	0.9835	0.6147
2.9559	2.625	105	4.3147	0.9322	0.5467	0.9231	0.1667	0.9048	0.9084	0.2047
0.0722	2.75	110	5.3917	0.9373	0.9991	0.8889	0.1333	0.8844	0.9701	0.4816
1.1721	2.875	115	5.6819	0.9402	0.9997	0.8803	0.1	0.8844	0.9778	0.5147
0.0083	3.0	120	3.6691	0.9368	0.9358	0.9231	0.1667	0.9048	0.9525	0.3350
0.1978	3.125	125	5.3790	0.9360	0.9971	0.9316	0.1667	0.9116	0.9702	0.4600
0.0148	3.25	130	7.5959	0.9363	1.0000	0.8718	0.1	0.8776	0.9801	0.4969
0.0006	3.375	135	6.2987	0.9259	0.9982	0.9231	0.2	0.8980	0.9640	0.3647
0.0002	3.5	140	6.6050	0.9219	0.9986	0.8974	0.2	0.8776	0.9609	0.3567
0.0011	3.625	145	7.8719	0.9148	0.9999	0.8974	0.2	0.8776	0.9665	0.3821
0.0148	3.75	150	8.5273	0.9080	1.0000	0.8803	0.1667	0.8707	0.9662	0.3833
0.0001	3.875	155	9.1221	0.8966	1.0000	0.8974	0.1667	0.8844	0.9655	0.3693
0.586	4.0	160	9.4509	0.9004	1.0000	0.8974	0.1667	0.8844	0.9665	0.3822
0.0001	4.125	165	9.3715	0.8990	1.0000	0.8974	0.1667	0.8844	0.9657	0.3700
0.0	4.25	170	9.1917	0.9105	1.0000	0.8974	0.1667	0.8844	0.9650	0.3646
0.0005	4.375	175	9.2575	0.8989	1.0000	0.8889	0.1667	0.8776	0.9653	0.3648
0.0	4.5	180	9.6203	0.8999	1.0000	0.8974	0.1667	0.8844	0.9656	0.3694
0.0002	4.625	185	9.8262	0.8989	1.0000	0.8974	0.1667	0.8844	0.9659	0.3729
0.0	4.75	190	9.8696	0.8991	1.0000	0.8974	0.1667	0.8844	0.9659	0.3782
0.0	4.875	195	9.6943	0.8996	1.0000	0.8974	0.1667	0.8844	0.9659	0.3762
0.0	5.0	200	9.7280	0.8987	1.0000	0.8974	0.1667	0.8844	0.9659	0.3777

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_2602261705

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo5_lr1e-4_r8_2602261705

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_2602261705