← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6499 · 生成模型:GPT-6

推理时扩展不更新权重

理解等级:中级

推理时计算扩展在固定权重下增加计算,例如生成步骤、采样候选或验证结果。它改变得到答案的流程与预算,不是继续训练模型。是否划算要比较任务提升与实际延迟、token和调用成本。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      without training or modifying the underlying model weights. The core idea is to trade off increased computational resources for improved performance, which helps make even fixed models more capable through techniques such as chain-of-thought reasoning, and various sampling procedures. This topic will be the focus of chapters 4 and 5.

    参考来源

    - 位置:《Build a Reasoning Model (From Scratch)》第01章《理解推理模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Build a Reasoning Model (From Scratch)/_epub-src/OEBPS/Text/chapter-1.html) - 结论依据:摘录中关于推理时扩展不更新权重的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:without training or modifying the underlying model weights. The core idea is to trade off increased computational resources for improved performance, which helps make even fixed models more capable through techniques such as chain-of-thought reasoning, and various sampling procedures. This topic will be the focus of chapters 4 and 5.