metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: source
dtype: string
- name: similarity_sequencematcher
dtype: float64
- name: most_similar_example_sequencematcher
dtype: string
- name: similarity_edit
dtype: int64
splits:
- name: train
num_bytes: 79185.40272028386
num_examples: 186
download_size: 56949
dataset_size: 79185.40272028386
configs:
- config_name: default
data_files:
- split: train
path: data/train-*