metadata
dataset_info:
features:
- name: text
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 654902
num_examples: 672
- name: test
num_bytes: 162814
num_examples: 168
download_size: 97425
dataset_size: 817716
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*