metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: initial_config
dtype: string
- name: path
sequence: string
- name: involved_classes
sequence: string
- name: answer
dtype: string
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
- name: user_question_bank
dtype: string
splits:
- name: train
num_bytes: 4078328
num_examples: 200
download_size: 288219
dataset_size: 4078328
configs:
- config_name: default
data_files:
- split: train
path: data/train-*