metadata
dataset_info:
features:
- name: file_id
dtype: string
- name: text_sub_id
dtype: int64
- name: text
dtype: string
- name: tokens
dtype: int64
splits:
- name: train
num_bytes: 1696432957
num_examples: 393386
download_size: 1069041271
dataset_size: 1696432957
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- en
English books from gutenberg.org with fiction tag and at least 25 downloads, split into paragraphs.
For license details see: https://www.gutenberg.org/policy/permission.html