metadata
dataset_info:
features:
- name: text
dtype: string
- name: perplexity
dtype: float64
splits:
- name: train
num_bytes: 609561256
num_examples: 186995
download_size: 290872095
dataset_size: 609561256
configs:
- config_name: default
data_files:
- split: train
path: data/train-*