metadata
dataset_info:
features:
- name: messages
list:
- name: from
dtype: string
- name: value
dtype: string
- name: label
dtype: bool
splits:
- name: train
num_bytes: 10504919
num_examples: 30000
download_size: 6646460
dataset_size: 10504919
configs:
- config_name: default
data_files:
- split: train
path: data/train-*