metadata
dataset_info:
features:
- name: id
dtype: string
- name: content
dtype: string
- name: max_stars_repo_path
dtype: string
splits:
- name: train
num_bytes: 1635647
num_examples: 199
download_size: 460021
dataset_size: 1635647
configs:
- config_name: default
data_files:
- split: train
path: data/train-*