metadata
dataset_info:
features:
- name: conversations
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 73810
num_examples: 123
- name: validation
num_bytes: 8407
num_examples: 13
download_size: 10688
dataset_size: 82217
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*