metadata
dataset_info:
features:
- name: input_ids
list: int32
- name: text
dtype: string
- name: source
dtype: string
- name: num_tokens
dtype: int32
splits:
- name: train
num_bytes: 68734031
num_examples: 16427
- name: validation
num_bytes: 6924886
num_examples: 1640
download_size: 61459934
dataset_size: 75658917
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*