metadata
dataset_info:
features:
- name: Human
dtype: string
- name: Assistant
dtype: string
- name: answer
dtype: string
- name: start_position
dtype: int64
- name: end_position
dtype: int64
- name: text
dtype: string
splits:
- name: train
num_bytes: 9332
num_examples: 30
download_size: 10023
dataset_size: 9332
configs:
- config_name: default
data_files:
- split: train
path: data/train-*