metadata
dataset_info:
features:
- name: incorrect
dtype: string
- name: correct
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 1196972
num_examples: 4935
download_size: 528921
dataset_size: 1196972
configs:
- config_name: default
data_files:
- split: train
path: data/train-*