Base Dataset: wikimedia/wikipedia (https://huggingface.co/datasets/wikimedia/wikipedia)
Extract only Korean data
license: cc-by-nc-sa-3.0 language:
- ko
dataset_info:
features:
- name: title dtype: string
- name: text dtype: string splits:
- name: train num_bytes: 1213180436.2364573 num_examples: 583107
- name: test num_bytes: 134798519.76354265 num_examples: 64790 download_size: 799654121 dataset_size: 1347978956.0 configs:
- config_name: default
data_files:
- split: train path: data/train-*
- split: test path: data/test-*