metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: label
dtype: string
- name: options
dtype: string
- name: predicted_label
dtype: string
- name: confidence
dtype: float64
- name: is_correct
dtype: bool
splits:
- name: train
num_bytes: 9025140
num_examples: 39720
download_size: 912476
dataset_size: 9025140
configs:
- config_name: default
data_files:
- split: train
path: data/train-*