metadata
dataset_info:
features:
- name: conversation
list:
- name: content
dtype: string
- name: role
dtype: string
- name: qa
struct:
- name: answer
dtype: string
- name: question
dtype: string
splits:
- name: train
num_bytes: 416599
num_examples: 500
download_size: 227462
dataset_size: 416599
configs:
- config_name: default
data_files:
- split: train
path: data/train-*