metadata
dataset_info:
features:
- name: text
dtype: string
- name: pile_idx
dtype: int64
splits:
- name: train
num_bytes: 5838447801
num_examples: 936630
- name: validation
num_bytes: 56650639
num_examples: 9366
- name: test
num_bytes: 2634699
num_examples: 469
download_size: 3247405148
dataset_size: 5897733139
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*