---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: metadata
    dtype: string
  splits:
  - name: train
    num_bytes: 44266046418
    num_examples: 2875658
  download_size: 13512248530
  dataset_size: 44266046418
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---