metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
- name: rationale
dtype: string
- name: config
dtype: string
- name: prompt
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 46085669
num_examples: 46382
- name: validation
num_bytes: 15267023
num_examples: 15461
- name: test
num_bytes: 15290221
num_examples: 15461
download_size: 44743821
dataset_size: 76642913
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*