metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: model_id
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 3009228
num_examples: 1248
download_size: 1982540
dataset_size: 3009228
configs:
- config_name: default
data_files:
- split: train
path: data/train-*