---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: token_type_ids
    list: int8
  - name: attention_mask
    list: int8
  splits:
  - name: train
    num_bytes: 16202271744
    num_examples: 5274177
  - name: test
    num_bytes: 382755840
    num_examples: 124595
  download_size: 4856526658
  dataset_size: 16585027584
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---