metadata
dataset_info:
features:
- name: conversation
list:
- name: name
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 4098113
num_examples: 11994
- name: valid
num_bytes: 1715097
num_examples: 4994
download_size: 3062673
dataset_size: 5813210
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*