modernbert-pii-mapped-v13

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: 0.0151
• Precision: 0.9599
• Recall: 0.9700
• F1: 0.9650
• Accuracy: 0.9966

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
0.1180	0.3322	500	0.0509	0.7814	0.8566	0.8173	0.9844
0.0591	0.6645	1000	0.0326	0.8853	0.9299	0.9070	0.9910
0.0383	0.9967	1500	0.0202	0.9219	0.9577	0.9395	0.9936
0.0359	1.3289	2000	0.0172	0.9408	0.9617	0.9511	0.9953
0.0233	1.6611	2500	0.0149	0.9469	0.9655	0.9561	0.9961
0.0226	1.9934	3000	0.0154	0.9458	0.9647	0.9551	0.9957
0.0137	2.3256	3500	0.0142	0.9583	0.9732	0.9657	0.9964
0.0112	2.6578	4000	0.0141	0.9542	0.9700	0.9620	0.9964
0.0105	2.9900	4500	0.0142	0.9523	0.9657	0.9589	0.9962
0.0049	3.3223	5000	0.0151	0.9599	0.9700	0.9650	0.9966

Framework versions

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