---
dataset_info:
  features:
  - name: tokens
    sequence: int64
  - name: tokens_reversed
    sequence: int64
  splits:
  - name: train
    num_bytes: 2959962112
    num_examples: 9999872
  download_size: 192990622
  dataset_size: 2959962112
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---