Datasets:

pmdlt
/

stem_wikipedia_chunked_selected

dataset_info:
  features:
    - name: doc_id
      dtype: int32
    - name: text
      dtype: string
    - name: source
      dtype: string
  splits:
    - name: train
      num_bytes: 48806512.416453354
      num_examples: 100000
  download_size: 42651977
  dataset_size: 48806512.416453354
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*