metadata
dataset_info:
features:
- name: id
dtype: string
- name: original_sentence
dtype: string
- name: inappropriate_part
dtype: string
- name: rewritten_part
dtype: string
- name: expected_score
dtype: float64
- name: description
dtype: string
splits:
- name: train
num_bytes: 8254938
num_examples: 35645
- name: validation
num_bytes: 946397
num_examples: 4036
- name: test
num_bytes: 2282322
num_examples: 9959
download_size: 5965560
dataset_size: 11483657
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*