metadata
dataset_info:
features:
- name: story
dtype: string
- name: generated_text_1
dtype: string
- name: generated_text_2
dtype: string
- name: mic_chosen
dtype: int64
- name: mar_chosen
dtype: int64
- name: ali_chosen
dtype: int64
- name: index
dtype: int64
- name: chosen
dtype: string
- name: rejected
dtype: string
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 75999.67741935483
num_examples: 65
- name: test
num_bytes: 32738.322580645163
num_examples: 28
download_size: 92284
dataset_size: 108738
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*