---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: token_type_ids
    sequence: int8
  - name: attention_mask
    sequence: int8
  - name: special_tokens_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 13366678800
    num_examples: 55694495
  - name: validation
    num_bytes: 143604720
    num_examples: 598353
  download_size: 5542068106
  dataset_size: 13510283520
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---