metadata
dataset_info:
features:
- name: examples
struct:
- name: conversations
sequence: string
- name: quality_score
dtype: float64
- name: source
dtype: string
splits:
- name: train
num_bytes: 425994
num_examples: 150
download_size: 249832
dataset_size: 425994
configs:
- config_name: default
data_files:
- split: train
path: data/train-*