metadata
dataset_info:
features:
- name: Conversation_no
dtype: int64
- name: Turn_no
dtype: int64
- name: Question
dtype: string
- name: Answer
dtype: string
- name: Topic
dtype: string
- name: Topic_section
dtype: string
- name: Rationale
dtype: string
- name: is_nq
dtype: bool
- name: Context
sequence: string
- name: Additional_answers
sequence: 'null'
splits:
- name: train
num_bytes: 56897630
num_examples: 45450
- name: validation
num_bytes: 3021520
num_examples: 2514
download_size: 16047738
dataset_size: 59919150
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*