metadata
dataset_info:
features:
- name: description
dtype: string
- name: prediction
dtype: string
splits:
- name: train
num_bytes: 787989
num_examples: 1989
download_size: 149164
dataset_size: 787989
configs:
- config_name: default
data_files:
- split: train
path: data/train-*