modernbert-pii-mapped-v10

This model is a fine-tuned version of answerdotai/ModernBERT-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0077
• Precision: 0.9759
• Recall: 0.9896
• F1: 0.9827
• Accuracy: 0.9981

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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_with_restarts
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0641	0.3591	500	0.0299	0.8870	0.9455	0.9153	0.9914
0.0396	0.7181	1000	0.0148	0.9440	0.9813	0.9623	0.9955
0.0178	1.0768	1500	0.0098	0.9615	0.9864	0.9738	0.9970
0.0154	1.4359	2000	0.0120	0.9617	0.9905	0.9759	0.9967
0.0156	1.7950	2500	0.0096	0.9615	0.9811	0.9712	0.9970
0.0064	2.1537	3000	0.0080	0.9707	0.9924	0.9814	0.9976
0.0063	2.5127	3500	0.0070	0.9749	0.9937	0.9842	0.9979
0.0081	2.8718	4000	0.0072	0.9737	0.9902	0.9819	0.9980
0.0020	3.2305	4500	0.0071	0.9762	0.9931	0.9846	0.9981
0.0036	3.5896	5000	0.0075	0.9739	0.9890	0.9814	0.9980
0.0018	3.9487	5500	0.0074	0.9747	0.9856	0.9801	0.9980
0.0007	4.3074	6000	0.0077	0.9759	0.9896	0.9827	0.9981

Framework versions

• Transformers 5.9.0
• Pytorch 2.12.0+cu130
• Datasets 4.8.5
• Tokenizers 0.22.2