metadata
dataset_info:
features:
- name: messages
dtype: string
splits:
- name: train
num_bytes: 4704
num_examples: 3
download_size: 11359
dataset_size: 4704
configs:
- config_name: default
data_files:
- split: train
path: data/train-*