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: 0.0726
• Auprc: 0.9955
• Auroc: 0.9975
• Tpr@1fpr: 0.9475

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
0.1567	0.1104	2000	0.1924	0.9640	0.9752	0.7752
0.1318	0.2207	4000	0.1906	0.9724	0.9844	0.7541
0.0919	0.3311	6000	0.0860	0.9911	0.9949	0.9060
0.0772	0.4414	8000	0.0791	0.9922	0.9954	0.9196
0.0899	0.5518	10000	0.0794	0.9927	0.9958	0.9222
0.0855	0.6621	12000	0.0807	0.9928	0.9959	0.9156
0.0738	0.7725	14000	0.0756	0.9935	0.9961	0.9280
0.0687	0.8829	16000	0.0734	0.9938	0.9964	0.9321
0.0618	0.9932	18000	0.0686	0.9940	0.9965	0.9328
0.0428	1.1036	20000	0.0747	0.9945	0.9968	0.9369
0.0367	1.2139	22000	0.0795	0.9946	0.9968	0.9394
0.0504	1.3243	24000	0.0697	0.9950	0.9971	0.9414
0.0395	1.4346	26000	0.0753	0.9947	0.9970	0.9351
0.0405	1.5450	28000	0.0776	0.9949	0.9970	0.9468
0.0365	1.6554	30000	0.0762	0.9951	0.9972	0.9402
0.0284	1.7657	32000	0.0734	0.9953	0.9974	0.9475
0.0268	1.8761	34000	0.0738	0.9954	0.9974	0.9468
0.0393	1.9864	36000	0.0726	0.9955	0.9975	0.9475

Framework versions

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