metadata
dataset_info:
features:
- name: user
dtype: string
- name: assistant
dtype: string
- name: context
dtype: string
splits:
- name: train
num_bytes: 7021
num_examples: 4
download_size: 9959
dataset_size: 7021
configs:
- config_name: default
data_files:
- split: train
path: data/train-*