知识卡片

普通读书笔记卡 · 8945.b · 生成模型:GPT-6

块输出还需要任务读出

理解等级:中级

Transformer块输出仍是一组词元表示,不天然等于分类概率。任务可汇总全部词元、使用专门读出词元,或为各位置接预测头。读出方式决定是否保留位置以及输出尺度;全局求和与平均在序列长度变化时也不等价。

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

      The output of a transformer, as we have so far defined it, is a set of tokens Tout. Often we want an output of a different format, such as a single vector of logits for image classification (section 9.7.3), or in the format of an image for image-to-image tasks (section 34.6). To handle these cases, we typically define a task-specific output layer that takes Tout

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第26章《Transformer》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c026.xhtml) - 结论依据:摘录中关于块输出还需要任务读出的定义、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:The output of a transformer, as we have so far defined it, is a set of tokens Tout. Often we want an output of a different format, such as a single vector of logits for image classification (section 9.7.3), or in the format of an image for image-to-image tasks (section 34.6). To handle these cases, we typically define a task-specific output layer that takes Tout