metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: requirements
dtype: string
- name: response_0
dtype: string
- name: response_1
dtype: string
- name: check
dtype: string
- name: response_0_1_judged_preference_majority
sequence: string
- name: response_0_1_judged_preference_mean
sequence: 'null'
splits:
- name: train
num_bytes: 3785
num_examples: 1
download_size: 29011
dataset_size: 3785
configs:
- config_name: default
data_files:
- split: train
path: data/train-*