---
dataset_info:
  features:
  - name: tokens
    sequence: int64
  splits:
  - name: train
    num_bytes: 91997587692
    num_examples: 7913769
  - name: validation
    num_bytes: 1163868088
    num_examples: 100000
  download_size: 17499918408
  dataset_size: 93161455780
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---