---
license: mit
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  splits:
  - name: train
    num_bytes: 35374431000.0
    num_examples: 8627910
  download_size: 17369358731
  dataset_size: 35374431000.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---