Flint, Recursive’s new 40B open model, can run interpretability experiments on itself!
摘要
Recursive 发布 40B 开源模型 Flint，该模型能对自身运行可解释性实验。它自主找到一个与守门员相关的神经元，并放大其输出方向，此后每条新回复都会提到“goalkeeper”。
摘要由模型根据原文撰写，以原文为准。
正文 · 原文
Flint, Recursive’s new 40B open model, can run interpretability experiments on itself!
It autonomously found a goalkeeper-related neuron, then amplified its output direction resulting in every new response mentioning “goalkeeper"
来源：alphaXiv · x.com
引用原文
askalphaxiv. Flint, Recursive’s new 40B open model, can run interpretability experiments on itself![EB/OL]. (2026-11-05)[访问日期]. https://x.com/askalphaxiv/status/2118353139598532938.