metadata
dataset_info:
features:
- name: messages
dtype: string
- name: metadata
dtype: string
splits:
- name: train
num_bytes: 57749897
num_examples: 5399
download_size: 11901885
dataset_size: 57749897
configs:
- config_name: default
data_files:
- split: train
path: data/train-*