metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: labels_params
struct:
- name: entity1
dtype: string
- name: entity2
dtype: string
- name: entity3
dtype: string
- name: entity4
dtype: string
splits:
- name: train
num_bytes: 31072
num_examples: 300
download_size: 9395
dataset_size: 31072
configs:
- config_name: default
data_files:
- split: train
path: data/train-*