metadata
dataset_info:
features:
- name: prompt_id
dtype: int64
- name: prompt
dtype: string
- name: example
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 166226656
num_examples: 80000
- name: validation
num_bytes: 20778332
num_examples: 10000
- name: test
num_bytes: 20778332
num_examples: 10000
download_size: 2017234
dataset_size: 207783320
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*