metadata
dataset_info:
features:
- name: Setting
dtype: string
- name: World
dtype: string
- name: num_train
dtype: int64
- name: train_data
list:
- name: question
dtype: string
- name: options
list: string
- name: correct_option
dtype: int64
- name: verification
struct:
- name: passed
dtype: bool
- name: errors
list: 'null'
- name: review
struct:
- name: question
struct:
- name: evidence
list: string
- name: reason
dtype: string
- name: valid
dtype: bool
- name: distractors
struct:
- name: reason
dtype: string
- name: valid
dtype: bool
- name: answers
list:
- name: reason
dtype: string
- name: valid
dtype: bool
- name: num_test
dtype: int64
- name: test_data
list:
- name: question
dtype: string
- name: options
list: string
- name: correct_option
dtype: int64
- name: verification
struct:
- name: passed
dtype: bool
- name: errors
list: 'null'
- name: review
struct:
- name: question
struct:
- name: evidence
list: string
- name: reason
dtype: string
- name: valid
dtype: bool
- name: distractors
struct:
- name: reason
dtype: string
- name: valid
dtype: bool
- name: answers
list:
- name: reason
dtype: string
- name: valid
dtype: bool
splits:
- name: train
num_bytes: 26159677
num_examples: 64
download_size: 25800163
dataset_size: 26159677
configs:
- config_name: default
data_files:
- split: train
path: data/train-*