metadata
dataset_info:
features:
- name: label
dtype: int64
- name: rating
dtype: string
- name: category
dtype: string
- name: rationale
dtype: string
splits:
- name: train
num_bytes: 1675715
num_examples: 3699
download_size: 460920
dataset_size: 1675715
configs:
- config_name: default
data_files:
- split: train
path: data/train-*