metadata
dataset_info:
features:
- name: article_id
dtype: string
- name: title
dtype: string
- name: url
dtype: string
- name: positive
dtype: string
- name: negatives
sequence: string
splits:
- name: train
num_bytes: 103891292
num_examples: 30280
download_size: 65669743
dataset_size: 103891292
configs:
- config_name: default
data_files:
- split: train
path: data/train-*