PrefixQuant

[Arxiv 2024] PrefixQuant: Static Quantization Beats Dynamic through Prefixed Outliers in LLMs
ContentsIntroductionMethodExperimentsReferencesIntroduction 作者提出 PrefixQuant，基于 QuaRot，通过在 WA 量化时
staticQuantizationPrefixQuantarxivOutliers
admin1月前
210