metadata
dataset_info:
features:
- name: formatted
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 4787
num_examples: 5
- name: validation
num_bytes: 1233
num_examples: 1
download_size: 11150
dataset_size: 6020
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*