---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 1748222697.2594833
    num_examples: 835900
  - name: eval
    num_bytes: 437057765.7405166
    num_examples: 208976
  download_size: 279998506
  dataset_size: 2185280463.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
---