metadata
dataset_info:
features:
- name: wrong
dtype: string
- name: corrected
dtype: string
splits:
- name: train
num_bytes: 380872861
num_examples: 938106
download_size: 246178720
dataset_size: 380872861
configs:
- config_name: default
data_files:
- split: train
path: data/train-*