metadata
dataset_info:
features:
- name: context
dtype: string
- name: response
dtype: string
splits:
- name: train
num_bytes: 3753161
num_examples: 23263
download_size: 1554612
dataset_size: 3753161
configs:
- config_name: default
data_files:
- split: train
path: data/train-*