Peter Steinberger 🦞@steipete · XAI 摘要
做角色评估时发现 Claude 总把自己排第一，于是移除评判中的模型名称并调整设置，避免模型自我偏好影响结果。
I'm working on character evals and noticed that Claude would constantly pick itself as #1， so I removed the model names from the judge and changed things.
Peter Steinberger 🦞@steipeteAI 摘要
做角色评估时发现 Claude 总把自己排第一，于是移除评判中的模型名称并调整设置，避免模型自我偏好影响结果。
I'm working on character evals and noticed that Claude would constantly pick itself as #1， so I removed the model names from the judge and changed things.
Peter Steinberger 🦞@steipete · X在 X 查看原推