---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  splits:
  - name: train
    num_bytes: 708311652
    num_examples: 1372697
  - name: valid
    num_bytes: 7259088
    num_examples: 14068
  download_size: 313854357
  dataset_size: 715570740
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
---