metadata
dataset_info:
features:
- name: title
dtype: string
- name: arxiv_id
dtype: string
splits:
- name: train
num_bytes: 33906
num_examples: 390
download_size: 21140
dataset_size: 33906
configs:
- config_name: default
data_files:
- split: train
path: data/train-*