metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 31573
num_examples: 45
- name: validation
num_bytes: 5006
num_examples: 8
download_size: 23859
dataset_size: 36579
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*