metadata
dataset_info:
features:
- name: contest
dtype: string
- name: contest_url
dtype: string
- name: url
dtype: string
- name: alphabet
dtype: string
- name: name
dtype: string
- name: score
dtype: string
- name: correct
dtype: int64
- name: total
dtype: int64
- name: editorials
list:
- name: content
dtype: string
- name: text
dtype: string
- name: url
dtype: string
- name: task_content
dtype: string
splits:
- name: train
num_bytes: 4432243
num_examples: 1971
download_size: 1994205
dataset_size: 4432243
configs:
- config_name: default
data_files:
- split: train
path: data/train-*