FrameByFrame/privacy-filter-korean

Token Classification·FrameByFrame· 125.6K· 2

apache-2.0 Token Classification 1.4B params base_model:openai/privacy-filter

Korean fine-tune of OpenAI Privacy Filter for span-level PII detection. Adapted via LoRA on attention projections only — the base's sparse-MoE backbone (1.5B / 50M active params) is kept frozen.

Open in MLForge Sign up free Desktop app Source ↗

# pull & run locally
pip install mlforge-sdk && mlforge pull FrameByFrame/privacy-filter-korean

Model details

Task

Token Classification

Provider

FrameByFrame

Parameters

1.4B

Size

2.6 GB

License

apache-2.0

Downloads

125.6K

Likes

Updated

2026-05-28

About FrameByFrame/privacy-filter-korean

Korean fine-tune of OpenAI Privacy Filter for span-level PII detection. Adapted via LoRA on attention projections only — the base's sparse-MoE backbone (1.5B / 50M active params) is kept frozen.

Related Token Classification

I w11wo/indonesian-roberta-base-posp-tagger Token Classification ·124.1M params 2.8M 10 🤗 HF B dslim/bert-base-NER Token Classification ·108.3M params 1.8M 718 🤗 HF S StanfordAIMI/stanford-deidentifier-base Token Classification 1.3M 81 🤗 HF W Babelscape/wikineural-multilingual-ner Token Classification ·177.3M params 889.1K 165 🤗 HF F oliverguhr/fullstop-punctuation-multilang-large Token Classification 648.8K 178 🤗 HF X Davlan/xlm-roberta-large-ner-hrl Token Classification ·558.9M params 639.9K 15 🤗 HF

Browse all Token Classification models →