metadata
dataset_info:
features:
- name: id
dtype: int64
- name: prompt
dtype: string
- name: score
dtype: int64
- name: embedding
sequence: float64
splits:
- name: train
num_bytes: 1387051325
num_examples: 109101
download_size: 1040902994
dataset_size: 1387051325
configs:
- config_name: default
data_files:
- split: train
path: data/train-*