---
dataset_info:
  features:
  - name: index
    dtype: int64
  - name: input
    dtype: string
  - name: expected
    dtype: string
  - name: prediction
    dtype: string
  - name: exact_match
    dtype: bool
  splits:
  - name: train
    num_bytes: 2893976
    num_examples: 614
  download_size: 659836
  dataset_size: 2893976
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---