metadata
dataset_info:
features:
- name: lang
dtype: string
- name: file_path
dtype: string
- name: repo_name
dtype: string
- name: commit
dtype: string
- name: file_code
dtype: string
- name: prefix
dtype: string
- name: suffix
dtype: string
- name: middle
dtype: string
- name: strategy
dtype: string
- name: context_items
list:
- name: content
dtype: string
- name: file_path
dtype: string
- name: rank
dtype: int64
- name: score
dtype: float64
splits:
- name: train
num_bytes: 1834820240
num_examples: 13489
download_size: 625796897
dataset_size: 1834820240
configs:
- config_name: default
data_files:
- split: train
path: data/train-*