metadata
dataset_info:
features:
- name: episode
dtype: string
- name: subtitle
dtype: string
- name: place
dtype: string
- name: person
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 566398
num_examples: 3619
download_size: 172612
dataset_size: 566398
configs:
- config_name: default
data_files:
- split: train
path: data/train-*