metadata
dataset_info:
features:
- name: title
dtype: string
- name: topics
sequence:
- name: posts
sequence:
- name: text
dtype: string
splits:
- name: train
num_bytes: 23170678001
num_examples: 875140
download_size: 11291631531
dataset_size: 23170678001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*