---
dataset_info:
  features:
  - name: sentence
    dtype: string
  - name: unsandhied
    dtype: string
  splits:
  - name: train
    num_bytes: 10889800
    num_examples: 98942
  - name: validation
    num_bytes: 470628
    num_examples: 4200
  - name: test
    num_bytes: 470628
    num_examples: 4200
  - name: test_500
    num_bytes: 58795
    num_examples: 500
  - name: validation_500
    num_bytes: 58795
    num_examples: 500
  download_size: 7399131
  dataset_size: 11948646
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
  - split: test_500
    path: data/test_500-*
  - split: validation_500
    path: data/validation_500-*
---