metadata
dataset_info:
features:
- name: name
dtype: string
- name: kind
dtype: string
- name: before
dtype: string
- name: after
dtype: string
- name: path
dtype: string
splits:
- name: train
num_bytes: 255693479
num_examples: 357280
download_size: 20504944
dataset_size: 255693479
configs:
- config_name: default
data_files:
- split: train
path: data/train-*