metadata
dataset_info:
features:
- name: question
dtype: string
- name: exp
dtype: string
- name: cop
dtype: int64
- name: opa
dtype: string
- name: opb
dtype: string
- name: opc
dtype: string
- name: opd
dtype: string
- name: subject_name
dtype: string
- name: topic_name
dtype: string
- name: id
dtype: string
- name: choice_type
dtype: string
- name: rewards
sequence: int64
- name: response
dtype: string
splits:
- name: train
num_bytes: 1343094959
num_examples: 199580
download_size: 570465702
dataset_size: 1343094959
configs:
- config_name: default
data_files:
- split: train
path: data/train-*