vnixxa31/commitpackft
Dataset Card for CommitPackFT Dataset Summary CommitPackFT is a 2GB filtered version of CommitPack to contain only high-quality commit messages that resemble natural language instructions. Creation: The dataset can be recreated using instructions available here. Languages: 277 OctoPack🐙🎒: Data CommitPack 4TB of GitHub commits across 350 programming languages CommitPackFT Filtered version of CommitPack for high-quality commit messages that resemble… See the full description on the dataset page: https://huggingface.co/datasets/vnixxa31/commitpackft.
156
1version https://git-lfs.github.com/spec/v12oid sha256:894bb373224dc45e2af6c69e224b2bd1b5b5c1e05e0abba5ce1f93e7e115464d3size 14402834 