metadata
dataset_info:
features:
- name: id
dtype: large_string
- name: submitter
dtype: large_string
- name: authors
dtype: large_string
- name: title
dtype: large_string
- name: comments
dtype: large_string
- name: journal-ref
dtype: large_string
- name: doi
dtype: large_string
- name: report-no
dtype: large_string
- name: categories
dtype: large_string
- name: license
dtype: large_string
- name: abstract
dtype: large_string
- name: versions
large_list:
- name: version
dtype: large_string
- name: created
dtype: large_string
- name: update_date
dtype: timestamp[ms]
- name: authors_parsed
large_list:
large_list: large_string
- name: prompt
dtype: large_string
- name: label
dtype: string
- name: prob
dtype: float64
splits:
- name: train
num_bytes: 717066848
num_examples: 213797
download_size: 375955571
dataset_size: 717066848
configs:
- config_name: default
data_files:
- split: train
path: data/train-*