metadata
dataset_info:
features:
- name: user_utterance
dtype: string
splits:
- name: train
num_bytes: 107205832
num_examples: 1772466
download_size: 66275655
dataset_size: 107205832
configs:
- config_name: default
data_files:
- split: train
path: data/train-*