论文73°
LLM内部状态预测研究
精选理由
Hermes 8B研究显示：用8个数字就能预测LLM内部状态走向，误差降低七成。
研究人员发现可通过8个数字预测LLM内部状态走向。在冻结的Hermes 8B模型中，从单层提取的8个数值可预测后续68个测量值。预测误差比简单基线模型低69-76%。
原文 · rohanpaul_ai
You can predict where an LLM's internal state is heading, and that targeted edits can pull it back on course.
On a frozen Hermes 8B, 8 numbers, taken from the model's hidden states at one layer, predict 68 measurements of what happens several layers later.
The prediction error comes out about 69-76% lower than a simple average baseline.
