Downloads · 30 days
6
2% of all-time downloads
pii-layout-synth-distil-v2
jefftherover/pii-layout-synth-distil-v2
pii-layout-synth-distil-v2 is a token classification model from jefftherover. Use it when you need labels on individual words, such as names. It is set up for transformers.
<a href="https://huggingface.co/spaces/jefftherover/huggingface-static-d56766" target="blank"<img src="https://raw.githubusercontent.com/gradio-app/trackio/refs/heads/main/trackio/assets/badge.png" alt="Visualize in T…
All-time downloads
374
Public
Parameters
94.5M
2.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors378 MB · 99%
At a glance
- Task
- Token Classification
- Library
- transformers
- Model type
- modernbert
- Access
- Public
- Created
- Jul 14, 2026
- Updated
- Jul 15, 2026
- SHA
- 3944e6c8
Try a prompt
Model card
From the Hugging Face model README
<a href="https://huggingface.co/spaces/jefftherover/huggingface-static-d56766" target="_blank"><img src="https://raw.githubusercontent.com/gradio-app/trackio/refs/heads/main/trackio/assets/badge.png" alt="Visualize in Trackio" title="Visualize in Trackio" style="height: 40px;"/></a>
<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->pii-layout-synth-distil-v2
This model is a fine-tuned version of jefftherover/pii-layout-synth-distil-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0048
- Precision: 0.9916
- Recall: 0.9951
- F1: 0.9933
- 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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|---|---|---|
| 9.4312 | 0.9685 | 2000 | 0.0066 | 0.9893 | 0.9944 | 0.9918 | 0.9984 |
| 8.3837 | 1.9370 | 4000 | 0.0056 | 0.9896 | 0.9935 | 0.9916 | 0.9986 |
| 7.6715 | 2.9056 | 6000 | 0.0049 | 0.9915 | 0.9948 | 0.9931 | 0.9987 |
| 7.4892 | 3.8741 | 8000 | 0.0048 | 0.9914 | 0.9947 | 0.9930 | 0.9988 |
| 7.1226 | 4.8426 | 10000 | 0.0048 | 0.9914 | 0.9949 | 0.9932 | 0.9988 |
| 7.3554 | 5.0 | 10325 | 0.0048 | 0.9916 | 0.9951 | 0.9933 | 0.9988 |
Framework versions
- Transformers 5.13.1
- Pytorch 2.13.0+cu130
- Datasets 5.0.0
- Tokenizers 0.22.2
- Task
- Token Classification
- Library
- transformers
- Type
- modernbert
- Created
- Jul 14, 2026
- Updated
- Jul 15, 2026