---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 5998238715
    num_examples: 2119719
  - name: validation
    num_bytes: 61703030
    num_examples: 21990
  download_size: 1771825341
  dataset_size: 6059941745
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---