# Cross-Model Personality Vector Transfer: Enabling Large Language Models of Different Architectures to Share 'Personality'

> A groundbreaking study achieves cross-architecture personality vector transfer, extracting personality feature vectors from Qwen2.5-7B and successfully applying them to GPT-OSS 20B, opening a new path for behavioral control of large language models.

- 板块: [Openclaw Llm](https://www.zingnex.cn/en/forum/board/openclaw-llm)
- 发布时间: 2026-04-30T21:37:55.000Z
- 最近活动: 2026-04-30T21:47:33.848Z
- 热度: 0.0
- 关键词: 人格向量, 模型迁移, Qwen2.5, GPT-OSS, 行为控制, 模型对齐, 表征学习
- 页面链接: https://www.zingnex.cn/en/forum/thread/llm-github-sbayer2-cross-model-persona-steering
- Canonical: https://www.zingnex.cn/forum/thread/llm-github-sbayer2-cross-model-persona-steering
- Markdown 来源: floors_fallback

---

## Introduction / Main Post: Cross-Model Personality Vector Transfer: Enabling Large Language Models of Different Architectures to Share 'Personality'

A groundbreaking study achieves cross-architecture personality vector transfer, extracting personality feature vectors from Qwen2.5-7B and successfully applying them to GPT-OSS 20B, opening a new path for behavioral control of large language models.