You can deploy this model and start earning money today!

leobianco/
npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo10_lr5e-4_r8_2602111128

Name: npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo10_lr5e-4_r8_2602111128
Availability: InStock

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

apache-2.0

Views

No views yet

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo10_lr5e-4_r8_2602111128

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: 5.1809
Roc Auc: 0.5107
Best Threshold: 0.5946
Tpr At Best Threshold: 0.3419
Fpr At Best Threshold: 0.2333
Accuracy At Best Threshold: 0.4286
Avg Score True Positives: 0.5547
Avg Score True Negatives: 0.5559

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: 0.0005
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: 10.0

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
75.0938	0.125	5	78.5227	0.4553	0.0402	0.5641	0.5	0.5510	0.4669	0.4723
97.0644	0.25	10	55.3650	0.4583	inf	0.0	0.0	0.2041	0.8500	0.8998
74.0028	0.375	15	89.9416	0.4252	0.0000	0.9316	0.9	0.7619	0.2587	0.3179
64.1465	0.5	20	50.4709	0.3996	inf	0.0	0.0	0.2041	0.7226	0.8477
30.8989	0.625	25	52.9709	0.4893	0.9999	0.8205	0.7667	0.7007	0.8744	0.9343
100.0156	0.75	30	165.3338	0.3936	0.0002	0.1453	0.1333	0.2925	0.0459	0.0994
97.4177	0.875	35	43.1330	0.4459	0.9702	0.9744	0.9	0.7959	0.9877	0.9424
37.7518	1.0	40	32.9967	0.5444	0.9962	0.4274	0.3	0.4830	0.5735	0.5301
19.7308	1.125	45	96.4766	0.5226	1.0000	0.1282	0.0333	0.2993	0.1816	0.1343
57.8131	1.25	50	36.9265	0.5010	1.0000	0.4872	0.4	0.5102	0.8002	0.8244
45.0414	1.375	55	32.9637	0.5634	1.0000	0.5983	0.4667	0.5850	0.8912	0.8266
7.9551	1.5	60	23.2751	0.7155	1.0000	0.4872	0.1	0.5714	0.7942	0.5345
30.7755	1.625	65	24.7064	0.7265	0.9850	0.7607	0.3667	0.7347	0.8422	0.4973
28.75	1.75	70	85.7765	0.7140	0.0000	0.8462	0.4	0.7959	0.1910	0.1183
17.75	1.875	75	41.9715	0.6627	1.0	0.9915	0.6667	0.8571	1.0000	0.8478
13.3442	2.0	80	22.7750	0.7556	1.0	0.9060	0.4333	0.8367	0.9999	0.6355
1.3395	2.125	85	18.4625	0.8192	1.0000	0.8803	0.3333	0.8367	0.9656	0.5000
34.2868	2.25	90	25.3798	0.7792	0.0000	0.9402	0.4667	0.8571	0.5930	0.2431
5.3908	2.375	95	12.0301	0.8064	0.9883	0.9573	0.4333	0.8776	0.9743	0.5850
0.0003	2.5	100	31.5852	0.6946	1.0000	0.9915	0.6	0.8707	1.0000	0.6900
13.8882	2.625	105	23.5238	0.7225	1.0	0.9744	0.5333	0.8707	0.9917	0.7309
26.1875	2.75	110	17.8690	0.8526	0.4610	0.6239	0.0333	0.6939	0.6028	0.0786
4.7188	2.875	115	34.8037	0.7254	1.0000	0.9573	0.5	0.8639	0.9719	0.6064
0.0	3.0	120	48.4166	0.7976	1.0	0.8462	0.2667	0.8231	0.8708	0.4000
0.0	3.125	125	42.4168	0.7806	1.0	0.9658	0.4	0.8912	0.9658	0.4969
40.4375	3.25	130	49.3392	0.6833	1.0	1.0	0.6333	0.8707	1.0	0.7006
8.125	3.375	135	19.5752	0.8295	1.0000	0.9060	0.2667	0.8707	0.9734	0.5358
0.0	3.5	140	28.2709	0.7504	1.0	0.9658	0.4667	0.8776	0.9833	0.6686
0.0	3.625	145	19.8037	0.8165	1.0000	0.9658	0.3333	0.9048	0.9736	0.5828
0.0	3.75	150	17.4799	0.8234	1.0000	0.9744	0.3333	0.9116	0.9997	0.6619
0.0	3.875	155	7.9205	0.9154	0.9999	0.9487	0.2	0.9184	0.9737	0.4928
10.9627	4.0	160	20.3790	0.8265	1.0	0.9829	0.3333	0.9184	1.0000	0.6875
9.6796	4.125	165	10.5775	0.8947	0.9870	0.8889	0.1667	0.8776	0.9171	0.2775
72.125	4.25	170	24.7787	0.7167	1.0	1.0	0.5667	0.8844	1.0	0.7391
1.3379	4.375	175	6.2850	0.9162	0.9874	0.8803	0.1333	0.8776	0.9337	0.2715
1.4243	4.5	180	16.2571	0.8600	1.0000	0.9402	0.2333	0.9048	0.9728	0.5299
0.0118	4.625	185	22.7038	0.7691	1.0000	0.9573	0.4333	0.8776	0.9852	0.5023
0.0	4.75	190	55.3299	0.7167	1.0	1.0	0.5667	0.8844	1.0	0.6202
0.0	4.875	195	38.6240	0.7605	1.0000	0.9573	0.4333	0.8776	0.9658	0.4755
28.7512	5.0	200	63.5380	0.7437	0.9921	0.7350	0.3	0.7279	0.7661	0.3630
29.9492	5.125	205	22.4687	0.5185	0.9199	0.6581	0.4667	0.6327	0.6921	0.6530
47.0022	5.25	210	73.4933	0.4648	0.0000	0.5128	0.4333	0.5238	0.0838	0.1466
11.3552	5.375	215	19.9364	0.4459	0.0001	0.9658	0.9333	0.7823	0.5420	0.5952
50.6094	5.5	220	13.4477	0.4228	0.8124	1.0	0.9667	0.8027	0.9874	0.9625
17.9704	5.625	225	15.5800	0.3806	0.9508	0.9744	0.9667	0.7823	0.9929	0.9972
12.4688	5.75	230	9.1966	0.3729	0.6339	0.9487	0.9333	0.7687	0.8857	0.9277
16.0171	5.875	235	9.2284	0.4870	0.3621	0.6410	0.4667	0.6190	0.4344	0.4578
7.9238	6.0	240	4.6451	0.5003	0.5407	0.8718	0.7333	0.7483	0.6768	0.6582
5.9688	6.125	245	6.3884	0.5563	0.3621	0.8376	0.7	0.7279	0.4228	0.4041
5.3027	6.25	250	5.5553	0.6142	0.4854	0.6667	0.4	0.6531	0.5013	0.4773
5.7227	6.375	255	6.7421	0.5923	0.3844	0.6325	0.3667	0.6327	0.3997	0.3868
5.8984	6.5	260	4.5190	0.5296	0.6192	0.7607	0.6	0.6871	0.6613	0.6617
6.4844	6.625	265	4.2043	0.4934	0.8187	0.3333	0.2667	0.4150	0.7981	0.8024
6.3672	6.75	270	9.2691	0.5097	0.2031	0.9573	0.8	0.8027	0.2551	0.2697
8.1953	6.875	275	4.2126	0.4886	0.7417	0.9145	0.8	0.7687	0.7831	0.7894
6.4043	7.0	280	9.7547	0.5464	0.2234	0.6923	0.5333	0.6463	0.2346	0.2539
5.1172	7.125	285	4.4172	0.4662	0.6938	0.6154	0.5333	0.5850	0.7053	0.7211
7.3438	7.25	290	5.6671	0.5003	0.4192	0.9231	0.8667	0.7619	0.5021	0.5133
5.8672	7.375	295	6.0176	0.4987	0.4340	0.6325	0.5333	0.5986	0.4709	0.4783
6.3984	7.5	300	4.2727	0.5131	0.8004	0.4786	0.3667	0.5102	0.7734	0.7776
4.5469	7.625	305	7.4415	0.5090	0.4158	0.3675	0.2333	0.4490	0.3678	0.3731
4.6035	7.75	310	5.8837	0.4972	0.5631	0.3248	0.2	0.4218	0.4918	0.5031
4.5283	7.875	315	4.2585	0.4923	0.8350	0.3590	0.2333	0.4422	0.7944	0.8007
6.3672	8.0	320	5.7766	0.4962	0.5010	0.4530	0.3667	0.4898	0.4921	0.4988
5.6914	8.125	325	5.7445	0.4900	0.5770	0.1880	0.0667	0.3401	0.4924	0.4997
5.3867	8.25	330	4.2554	0.4623	0.8219	0.1453	0.0667	0.3061	0.7599	0.7702
4.6406	8.375	335	5.3021	0.4634	0.4863	0.8376	0.7667	0.7143	0.5445	0.5602
6.082	8.5	340	7.4809	0.4548	0.4301	0.1538	0.1	0.3061	0.3546	0.3753
5.5664	8.625	345	5.1718	0.4688	0.6220	0.2479	0.1333	0.3741	0.5620	0.5754
5.166	8.75	350	4.4103	0.5048	0.7441	0.2222	0.1	0.3605	0.6851	0.6878
5.918	8.875	355	4.5396	0.5026	0.7074	0.2906	0.1667	0.4014	0.6582	0.6634
5.2461	9.0	360	5.7793	0.4925	0.5181	0.3419	0.2333	0.4286	0.4900	0.4975
4.875	9.125	365	6.0230	0.5031	0.5093	0.3077	0.2	0.4082	0.4652	0.4694
5.4102	9.25	370	6.2077	0.5058	0.5068	0.2821	0.1667	0.3946	0.4486	0.4524
4.957	9.375	375	5.9243	0.4944	0.5419	0.2393	0.1333	0.3673	0.4757	0.4842
5.1016	9.5	380	5.6067	0.5194	0.5737	0.2479	0.1	0.3810	0.5050	0.5043
5.4609	9.625	385	5.3131	0.5060	0.6039	0.2735	0.1	0.4014	0.5392	0.5405
5.207	9.75	390	5.2379	0.4966	0.5908	0.3248	0.2	0.4218	0.5495	0.5539
5.2285	9.875	395	5.2034	0.4979	0.6252	0.2222	0.1	0.3605	0.5532	0.5561
5.291	10.0	400	5.1809	0.5107	0.5946	0.3419	0.2333	0.4286	0.5547	0.5559

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_epo10_lr5e-4_r8_2602111128

npov_RM_mistralai_S12345_LLM_false_STRUCT_true_epo10_lr5e-4_r8_2602111128

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_epo10_lr5e-4_r8_2602111128