---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  splits:
  - name: train
    num_bytes: 1240868820
    num_examples: 18933
  - name: valid
    num_bytes: 81439404
    num_examples: 1243
  - name: test
    num_bytes: 71831840
    num_examples: 1096
  download_size: 262541715
  dataset_size: 1394140064
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---