metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: source
dtype: string
- name: score
dtype: int64
splits:
- name: train
num_bytes: 278830.3357271095
num_examples: 501
- name: test
num_bytes: 31166.664272890484
num_examples: 56
download_size: 47314
dataset_size: 309997
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*