pii-layout-synth-v12

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: 0.0041
• Precision: 0.9926
• Recall: 0.9954
• F1: 0.9940
• Accuracy: 0.9988

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: 200
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0340	0.1863	500	0.0140	0.9722	0.9818	0.9770	0.9960
0.0238	0.3726	1000	0.0104	0.9820	0.9888	0.9854	0.9970
0.0138	0.5589	1500	0.0077	0.9852	0.9919	0.9885	0.9977
0.0126	0.7452	2000	0.0063	0.9873	0.9910	0.9891	0.9980
0.0143	0.9314	2500	0.0049	0.9914	0.9954	0.9934	0.9986
0.0058	1.1177	3000	0.0042	0.9920	0.9959	0.9939	0.9988
0.0073	1.3040	3500	0.0064	0.9891	0.9933	0.9912	0.9981
0.0081	1.4903	4000	0.0068	0.9891	0.9917	0.9904	0.9981
0.0058	1.6766	4500	0.0037	0.9932	0.9965	0.9949	0.9989
0.0075	1.8629	5000	0.0040	0.9918	0.9957	0.9938	0.9988
0.0030	2.0492	5500	0.0033	0.9935	0.9952	0.9944	0.9990
0.0033	2.2355	6000	0.0041	0.9926	0.9954	0.9940	0.9988

Framework versions

• Transformers 5.12.1
• Pytorch 2.12.1+cu130
• Datasets 5.0.0
• Tokenizers 0.22.2