metadata
dataset_info:
features:
- name: reponame
dtype: string
- name: instruction
dtype: string
- name: full_repo_dump
dtype: string
- name: output_code
dtype: string
- name: repo_tree
dtype: string
splits:
- name: train
num_bytes: 30776349
num_examples: 1
download_size: 30533315
dataset_size: 30776349
configs:
- config_name: default
data_files:
- split: train
path: data/train-*