metadata
dataset_info:
features:
- name: text
dtype: string
- name: document_url
dtype: string
- name: source_url
dtype: string
splits:
- name: train
num_bytes: 12807216
num_examples: 19724
download_size: 4716484
dataset_size: 12807216
configs:
- config_name: default
data_files:
- split: train
path: data/train-*