metadata
dataset_info:
features:
- name: output
dtype: string
- name: prompt
dtype: string
- name: extracted_answer
dtype: string
- name: correct
dtype: int64
splits:
- name: train
num_bytes: 230873
num_examples: 30
download_size: 126595
dataset_size: 230873
configs:
- config_name: default
data_files:
- split: train
path: data/train-*