Datasets:

JPPOL-AI
/

eb-ner-temp-test

dataset_info:
  features:
    - name: tokens
      list: string
    - name: labels
      list: string
    - name: content_id
      dtype: int64
    - name: cms_publication
      dtype: string
  splits:
    - name: train
      num_bytes: 863443
      num_examples: 2731
  download_size: 216072
  dataset_size: 863443
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Test split from JPPOL-AI/eb-ner. Re-uploaded for llm annotation evaluation. MISC category removed.