metadata
dataset_info:
features:
- name: Story
dtype: string
- name: Question
dtype: string
- name: span_start
dtype: float64
- name: span_end
dtype: float64
- name: span_text
dtype: string
- name: Answer
dtype: string
splits:
- name: train
num_bytes: 111024407
num_examples: 66611
download_size: 66935230
dataset_size: 111024407
configs:
- config_name: default
data_files:
- split: train
path: data/train-*