metadata
language: en
dataset_info:
features:
- name: title
dtype: string
- name: text
dtype: string
- name: reddit_scores
sequence: int32
splits:
- name: train
num_bytes: 68786199155
num_examples: 17103059
download_size: 42444568964
dataset_size: 68786199155