---
dataset_info:
  features:
  - name: sequence
    dtype: string
  splits:
  - name: train
    num_bytes: 19850004390.269722
    num_examples: 68340540
  - name: valid
    num_bytes: 2904572.365139304
    num_examples: 10000
  - name: test
    num_bytes: 2904572.365139304
    num_examples: 10000
  download_size: 19881625108
  dataset_size: 19855813535.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---