redline-guard-cap100k

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: 0.1093
• Auprc: 0.9916
• Auroc: 0.9954
• Tpr@1fpr: 0.9322

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
0.1527	0.1561	2000	0.1273	0.9793	0.9884	0.8557
0.1179	0.3121	4000	0.1310	0.9786	0.9861	0.9034
0.0973	0.4682	6000	0.1091	0.9886	0.9941	0.9019
0.0878	0.6243	8000	0.0927	0.9889	0.9931	0.9356
0.0809	0.7803	10000	0.0785	0.9907	0.9944	0.9398
0.0883	0.9364	12000	0.0803	0.9905	0.9945	0.9319
0.0451	1.0925	14000	0.1237	0.9900	0.9951	0.8921
0.0505	1.2485	16000	0.1267	0.9895	0.9942	0.9132
0.0404	1.4046	18000	0.1216	0.9906	0.9946	0.9270
0.0488	1.5607	20000	0.1056	0.9919	0.9959	0.9243
0.0411	1.7167	22000	0.1096	0.9919	0.9957	0.9240
0.0368	1.8728	24000	0.1093	0.9916	0.9954	0.9322

Framework versions

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