metadata
dataset_info:
features:
- name: text
dtype: string
- name: num_tokens
dtype: int64
- name: completion
dtype: string
- name: lyrics
dtype: string
splits:
- name: train
num_bytes: 7495757
num_examples: 7319
download_size: 3850201
dataset_size: 7495757
configs:
- config_name: default
data_files:
- split: train
path: data/train-*