metadata
dataset_info:
features:
- name: arxiv_id
dtype: string
- name: published_date
dtype: string
- name: categories
sequence: string
- name: title
dtype: string
- name: primary_category
dtype: string
- name: authors
sequence: string
- name: github
dtype: string
- name: is_github_official
dtype: bool
splits:
- name: train
num_bytes: 22157511
num_examples: 87987
download_size: 12702096
dataset_size: 22157511
configs:
- config_name: default
data_files:
- split: train
path: data/train-*