metadata
dataset_info:
features:
- name: title
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 1204098480
num_examples: 2524443
download_size: 687559046
dataset_size: 1204098480
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- en
Text split version of level 5 vital wikipedia articles. Text has been split on \n\n, and any instances where the length of the text was < 10 words were removed to avoid headings.