---
dataset_info:
  features:
  - name: article_id
    dtype: string
  - name: begins
    dtype: string
  - name: title
    dtype: string
  - name: type
    dtype: string
  - name: page
    dtype: int64
  - name: source_file
    dtype: string
  - name: newspaper_name
    dtype: string
  - name: date
    dtype: timestamp[us]
  - name: clean_ocr_text
    dtype: string
  - name: article_url
    dtype: string
  splits:
  - name: train
    num_bytes: 6170021244
    num_examples: 2618671
  download_size: 3131885489
  dataset_size: 6170021244
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---