---
dataset_info:
  features:
  - name: sequence
    dtype: string
  splits:
  - name: train
    num_bytes: 26063849199.91452
    num_examples: 84987726
  - name: valid
    num_bytes: 3066778.0427393154
    num_examples: 10000
  - name: test
    num_bytes: 3066778.0427393154
    num_examples: 10000
  download_size: 26147724298
  dataset_size: 26069982756.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---