metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: question
dtype: string
- name: options
list: string
- name: human_dist
list: float64
- name: answer
dtype: string
- name: dataset_name
dtype: string
- name: instrument
dtype: string
- name: treatment
dtype: string
- name: question_id
dtype: string
- name: outcome
dtype: string
- name: slice_n
dtype: int64
splits:
- name: train
num_bytes: 23154
num_examples: 5
- name: test
num_bytes: 36148
num_examples: 32
download_size: 36308
dataset_size: 59302
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*