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.
166
1version https://git-lfs.github.com/spec/v12oid sha256:f7d8c2b0a89f307d92fe68cf5a55da0cf2c5b53dc39fa8103d395d01a6c13c383size 30505484 