---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: validation
    num_bytes: 1366446672
    num_examples: 33354
  - name: train
    num_bytes: 135315747216
    num_examples: 3302962
  download_size: 53030280443
  dataset_size: 136682193888
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/validation-*
  - split: train
    path: data/train-*
---