⚡Megatron LLM Hub
←Back to search
Twitter

开源 UpSafe 用两阶段 upcycling 向 LLM 各层注入可控安全专家,并附推理期安全路由与越狱评测。

本地 LLM 加固的可复现方案,含越狱评估,可直接上手比对效果。
llm工具越狱
UpSafe — Research code for adding controllable safety experts to LLM layers via two-stage upcycling training, with inference-time safety temperature routing and jailbreak evaluation. https://t.co/wxzGUl41rM https://t.co/LNFXInpBVl