metadata
dataset_info:
features:
- name: dataset_id
dtype: string
- name: original_text
dtype: string
- name: rewritten_text
dtype: string
- name: prompt
dtype: string
- name: tune_response
dtype: string
splits:
- name: train
num_bytes: 18488880
num_examples: 11648
download_size: 11723193
dataset_size: 18488880
configs:
- config_name: default
data_files:
- split: train
path: data/train-*