---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  splits:
  - name: train
    num_bytes: 707172840
    num_examples: 1370490
  - name: valid
    num_bytes: 7148664
    num_examples: 13854
  download_size: 313437276
  dataset_size: 714321504
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
---