metadata
dataset_info:
features:
- name: title
dtype: string
- name: id
dtype: string
- name: url
dtype: string
- name: created_utc
dtype: timestamp[ns]
- name: author
dtype: string
- name: subreddit
dtype: string
- name: selftext
dtype: string
- name: score
dtype: int64
- name: num_comments
dtype: int64
- name: text_content
dtype: string
- name: comments_id
dtype: string
- name: comment_author
dtype: string
- name: comment_author_fullname
dtype: string
- name: comment_content
dtype: string
- name: distinguished
dtype: 'null'
- name: comment_score
dtype: int64
splits:
- name: train
num_bytes: 2853000538
num_examples: 256237
download_size: 76943606
dataset_size: 2853000538
configs:
- config_name: default
data_files:
- split: train
path: data/train-*