---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: source
    dtype: string
  - name: word_start
    dtype: int64
  - name: word_end
    dtype: int64
  splits:
  - name: train
    num_bytes: 355442793
    num_examples: 120385
  download_size: 168325467
  dataset_size: 355442793
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---