metadata
dataset_info:
features:
- name: messages
dtype: string
splits:
- name: train
num_bytes: 2840513
num_examples: 382
download_size: 803682
dataset_size: 2840513
configs:
- config_name: default
data_files:
- split: train
path: data/train-*