redline-guard

This model is a fine-tuned version of answerdotai/ModernBERT-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0493
• Auprc: 0.9979
• Auroc: 0.9985
• Tpr@1fpr: 0.9743

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 128
• seed: 42
• 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: linear
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Auprc	Auroc	Tpr@1fpr
0.149	0.0770	2000	0.1499	0.9800	0.9861	0.8161
0.0901	0.1541	4000	0.0956	0.9914	0.9941	0.9168
0.0743	0.2311	6000	0.0808	0.9940	0.9959	0.9348
0.0703	0.3082	8000	0.1379	0.9910	0.9943	0.8894
0.069	0.3852	10000	0.0690	0.9953	0.9969	0.9440
0.0589	0.4623	12000	0.0752	0.9954	0.9971	0.9407
0.0658	0.5393	14000	0.0744	0.9959	0.9972	0.9504
0.0615	0.6164	16000	0.0592	0.9963	0.9975	0.9598
0.0573	0.6934	18000	0.0642	0.9961	0.9974	0.9521
0.0512	0.7704	20000	0.0529	0.9970	0.9980	0.9666
0.0501	0.8475	22000	0.0494	0.9972	0.9981	0.9645
0.0573	0.9245	24000	0.0496	0.9972	0.9982	0.9656
0.0473	1.0016	26000	0.0494	0.9975	0.9983	0.9710
0.031	1.0786	28000	0.0561	0.9972	0.9981	0.9691
0.0314	1.1557	30000	0.0537	0.9974	0.9982	0.9704
0.0266	1.2327	32000	0.0574	0.9973	0.9982	0.9693
0.0286	1.3098	34000	0.0538	0.9974	0.9981	0.9713
0.0317	1.3868	36000	0.0554	0.9976	0.9983	0.9724
0.0281	1.4638	38000	0.0514	0.9976	0.9983	0.9737
0.0254	1.5409	40000	0.0578	0.9974	0.9982	0.9703
0.0287	1.6179	42000	0.0536	0.9976	0.9983	0.9726
0.0233	1.6950	44000	0.0514	0.9978	0.9985	0.9752
0.0215	1.7720	46000	0.0533	0.9976	0.9983	0.9751
0.031	1.8491	48000	0.0512	0.9978	0.9985	0.9740
0.0265	1.9261	50000	0.0493	0.9979	0.9985	0.9743

Framework versions

• Transformers 4.57.6
• Pytorch 2.13.0+cu130
• Datasets 5.0.1
• Tokenizers 0.22.2