metadata
dataset_info:
features:
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 2662215061
num_examples: 9335000
download_size: 413174374
dataset_size: 2662215061
configs:
- config_name: default
data_files:
- split: train
path: data/train-*