metadata
dataset_info:
features:
- name: idx
dtype: int64
- name: answer
dtype: string
- name: task
dtype: string
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 1559779
num_examples: 2334
download_size: 431119
dataset_size: 1559779
configs:
- config_name: default
data_files:
- split: train
path: data/train-*