metadata
dataset_info:
features:
- name: good_text
dtype: string
- name: bad_text
dtype: string
- name: error_family
dtype: string
- name: error_type
dtype: string
- name: uid
dtype: string
- name: misc
dtype: string
- name: error_class
dtype: int64
- name: source
dtype: string
splits:
- name: train
num_bytes: 51256221
num_examples: 77807
download_size: 24586018
dataset_size: 51256221
configs:
- config_name: default
data_files:
- split: train
path: data/train-*