metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: id
dtype: string
- name: prompt_educational_score
dtype: float64
- name: response_educational_score
dtype: float64
- name: average_educational_score
dtype: float64
splits:
- name: train
num_bytes: 47853064
num_examples: 22608
download_size: 20144400
dataset_size: 47853064
configs:
- config_name: default
data_files:
- split: train
path: data/train-*