metadata
dataset_info:
features:
- name: arxiv_id
dtype: string
- name: email
dtype: string
- name: title
dtype: string
- name: hf_original
sequence: string
- name: hf_baseline
sequence: string
- name: date
dtype: string
- name: mentions_github
sequence: string
- name: github_url
dtype: string
splits:
- name: train
num_bytes: 282360
num_examples: 891
download_size: 126465
dataset_size: 282360
configs:
- config_name: default
data_files:
- split: train
path: data/train-*