技术博客/Persona vectors: Monitoring and controlling character traits in language models
Anthropic进阶2025-08-01· 9 分钟· RAG 与记忆

Persona vectors: Monitoring and controlling character traits in language models

Interpretability Aug 1, 2025 Read the paper Language models are strange beasts. In many ways they appear to have human-like “personalities” and “moods

本文需解锁后阅读

免费开放 3 篇精选工程博客(列表置顶)。其余文章登录后可阅读全文。

人格向量:监测并控制语言模型中的性格特质 — AI Learner