metadata
dataset_info:
features:
- name: instruction_seed
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 30472947
num_examples: 100001
download_size: 15560223
dataset_size: 30472947
configs:
- config_name: default
data_files:
- split: train
path: data/train-*