metadata
dataset_info:
features:
- name: example_id
dtype: string
- name: article
dtype: string
- name: answer
dtype: string
- name: question
dtype: string
- name: options
sequence: string
- name: input_ids
sequence:
sequence: int32
- name: attention_mask
sequence:
sequence: int8
- name: label
dtype: int64
splits:
- name: train
num_bytes: 9381086
num_examples: 1436
download_size: 1252839
dataset_size: 9381086
configs:
- config_name: default
data_files:
- split: train
path: data/train-*