metadata
dataset_info:
features:
- name: title
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 416622171
num_examples: 1167
download_size: 252088459
dataset_size: 416622171
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- nl
Dutch Gutenberg Books
This dataset contains all 1167 Dutch books, gathered from the Project Gutenberg on 18 September 2026.
The dataset is suitable for language model pretraining, including continued pretraining (CPT).
License
Project Gutenberg states that the vast majority of its eBooks are in the public domain in the United States. This permits uses such as commercial use, republication, and the creation of derivative works.
Copyright status may differ depending on the individual work and the jurisdiction. Users should verify the applicable copyright status before using or redistributing the contents of individual books.
For more information, see Project Gutenberg's licensing information.