adam_karvonen· @adam_karvonen · X·· 6 小时前AI 评分
Astra 消融 Qwen 权重中的循环方向
看起来 Astra 并没有原生读取 Qwen 的权重？ 据我理解，Qwen 会陷入循环，无法回答问题。 Astra 使用代码执行了一些标准的可解释性技术，发现了一个循环方向，并将其从权重中消融。
It looks like Astra didn't natively read Qwen's weights?
As I understand it, Qwen would get stuck in a loop and not answer the question.
Astra used code to perform some standard interp techniques, found a looping direction, and ablated it from the weights.
来源：adam_karvonen · x.com