metadata
dataset_info:
features:
- name: sentence_good
dtype: string
- name: sentence_bad
dtype: string
- name: feature_name
dtype: string
splits:
- name: train
num_bytes: 5482428
num_examples: 46000
download_size: 1161379
dataset_size: 5482428
configs:
- config_name: default
data_files:
- split: train
path: data/train-*