metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completions
list: string
- name: labels
list: float64
splits:
- name: train_1
num_bytes: 14534690
num_examples: 4800
- name: train_2
num_bytes: 14230942
num_examples: 4800
- name: test
num_bytes: 4811536
num_examples: 2192
download_size: 33068710
dataset_size: 33577168
configs:
- config_name: default
data_files:
- split: train_1
path: data/train_1-*
- split: train_2
path: data/train_2-*
- split: test
path: data/test-*