metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: split
dtype: string
- name: source_dataset
dtype: string
- name: subject
dtype: string
splits:
- name: train
num_bytes: 258799210
num_examples: 151767
download_size: 119175306
dataset_size: 258799210
configs:
- config_name: default
data_files:
- split: train
path: data/train-*