metadata
dataset_info:
features:
- name: paper_url
dtype: string
- name: comment
dtype: string
splits:
- name: train
num_bytes: 501512
num_examples: 455
download_size: 128225
dataset_size: 501512
configs:
- config_name: default
data_files:
- split: train
path: data/train-*