---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: token_type_ids
    list: int8
  - name: attention_mask
    list: int8
  splits:
  - name: train
    num_bytes: 14248461312
    num_examples: 4638171
  - name: test
    num_bytes: 336356352
    num_examples: 109491
  download_size: 3986584006
  dataset_size: 14584817664
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---