metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: source
dtype: string
- name: score
dtype: float64
splits:
- name: train
num_bytes: 256843258.71779
num_examples: 99523
download_size: 116144945
dataset_size: 256843258.71779
configs:
- config_name: default
data_files:
- split: train
path: data/train-*