---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: special_tokens_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 7023275388.332596
    num_examples: 2277342
  - name: validation
    num_bytes: 372257304.0
    num_examples: 120706
  download_size: 2053364016
  dataset_size: 7395532692.332596
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---