metadata
dataset_info:
features:
- name: parent_url
dtype: string
- name: parent_score
dtype: string
- name: parent_body
dtype: string
- name: parent_user
dtype: string
- name: parent_title
dtype: string
- name: body
dtype: string
- name: score
dtype: string
- name: user
dtype: string
- name: answer_id
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 160950743
num_examples: 62556
download_size: 90399556
dataset_size: 160950743
configs:
- config_name: default
data_files:
- split: train
path: data/train-*