metadata
dataset_info:
features:
- name: instruction_seed
dtype: string
- name: source
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 40922892
num_examples: 150000
download_size: 8162888
dataset_size: 40922892
configs:
- config_name: default
data_files:
- split: train
path: data/train-*