← Back to feed

Shaping and Monitoring Models Through Their Internal Representations

AI · Studies · 2026-11-06 04:59

Goodfire发布关于大语言模型机制性塑造的研究，探讨如何通过干预模型内部表示来调整其行为与性能。

Source: www.goodfire.com

Auto-generated summary for reference only. RSS