metadata
dataset_info:
features:
- name: content
dtype: string
- name: sha1
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 20762952.644877758
num_examples: 37340
download_size: 11554736
dataset_size: 20762952.644877758
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
Subset of StarCoder2-Instruct's seed dataset. Used for experimentation.
Full dataset is found here: https://huggingface.co/datasets/bigcode/python-stack-v1-functions-filtered-sc2