metadata
dataset_info:
features:
- name: id
dtype: int64
- name: prompt
dtype: string
- name: label
dtype: string
- name: target_functions
list: string
- name: grading_function
dtype: string
- name: detailed_prompt
dtype: bool
- name: deceptive
dtype: bool
- name: hint_included
dtype: bool
splits:
- name: train
num_bytes: 86096
num_examples: 164
download_size: 30480
dataset_size: 86096
configs:
- config_name: default
data_files:
- split: train
path: data/train-*