metadata
dataset_info:
features:
- name: dialog1
dtype: string
- name: dialog2
dtype: string
splits:
- name: train
num_bytes: 287370189.9883655
num_examples: 774522
- name: test
num_bytes: 71842733.01163453
num_examples: 193631
download_size: 196001427
dataset_size: 359212923
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*