metadata
language:
- en
dataset_info:
features:
- name: sentence
dtype: string
- name: loss
dtype: float64
- name: groundtruth
dtype: int64
- name: gpt4_label
dtype: int64
- name: augmentedSentence
dtype: string
splits:
- name: train
num_bytes: 92929
num_examples: 1000
download_size: 54967
dataset_size: 92929
configs:
- config_name: default
data_files:
- split: train
path: data/train-*