metadata
dataset_info:
features:
- name: conversation
list:
- name: content
dtype: string
- name: role
dtype: string
- name: primary_tag
dtype: string
splits:
- name: train
num_bytes: 252675.99964955932
num_examples: 219
- name: val
num_bytes: 24268.188002649258
num_examples: 21
download_size: 14605
dataset_size: 276944.1876522086
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*