metadata
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: prompt_without_context
dtype: string
- name: prompt_with_context
dtype: string
- name: ground_truth
dtype: string
splits:
- name: train
num_bytes: 13872743
num_examples: 1192
download_size: 2288825
dataset_size: 13872743
configs:
- config_name: default
data_files:
- split: train
path: data/train-*