metadata
dataset_info:
features:
- name: split
dtype: string
- name: document_id
dtype: string
- name: questions
sequence: string
- name: options
sequence:
sequence: string
- name: gold_label
sequence: int64
- name: difficult
sequence: int64
- name: text
sequence:
sequence: string
splits:
- name: train
num_bytes: 4999089
num_examples: 190
download_size: 1780854
dataset_size: 4999089
configs:
- config_name: default
data_files:
- split: train
path: data/train-*