⚡Megatron LLM Hub
←Back to search
Twitter

Phantom-KV 通过加载式 KV cache 嫁接移除模型拒答,不改动权重。

无需微调即可逆改变本地模型拒答行为。
llmai_agent工具
RT @lordx64: I built a tool to control your models refusals from the KV cache without modifying the weights. Handy for Cyber capable models Phantom-KV: Refusal removal as a loadable KV-cache graft zero weight modification, no refusal-direction projection, fully reversible. https://t.co/KutyA17d2T