metadata
dataset_info:
features:
- name: article
dtype: string
- name: highlights
dtype: string
splits:
- name: train
num_bytes: 4296661
num_examples: 1000
download_size: 2587008
dataset_size: 4296661
configs:
- config_name: default
data_files:
- split: train
path: data/train-*