集里怎么说它

• 《把训练好的模型当数据喂给模型：权重空间学习》(14:16起):本集说窗口化是把模型参数当作长序列、只重建每个窗口的技巧，这解耦了原始模型和自编码器的序列对应关系，使方法能从玩具模型扩展到 ResNet 级别

① 提到它的金句

12 条

那种上下文无法进入人脑，但它可以进入 AI 的上下文窗口。
That context can’t go into a human brain, but it can go into the AI context window.
—— Evan Reiser · [08:04]

指向原始笔记的链接

所以请注意，如果你是一个创业公司甚至是 Salesforce 或 HubSpot，也许在未来 12 个月内关闭这些交易，因为窗口可能会关闭，人们会说，“听着，那是我见过的最酷的智能体。我对之前的五个感到筋疲力尽。
And so just be aware, if you’re a startup or even Salesforce or HubSpot, maybe close those deals in the next 12 months, because the window may close where people say, “Listen, that’s the coolest agent I’ve ever seen. I’m exhausted from the last five.
—— Jason Lemkin · [41:43]

指向原始笔记的链接

就像实习生一样，如果简单的事情在大多数情况下足够了，他们就不会去做难事。如果窗户开着，就不要去防锁，对吧？
Like interns, they’re not going to do the hard thing if the easy thing will suffice in most cases. Don’t proof the lock if the window’s open, right?
—— Aaron Zolman · [10:00]

指向原始笔记的链接

我们最初的尝试是把 Base64 数据直接传进代码，那是个非常糟糕的主意。它直接把上下文窗口撑爆了，方方面面都很糟。别那么做。
Our first attempt was just passing Base64 data into the code, and that was just a terrible idea. It just blew up the context window and was bad all around. Don’t do that.
—— Jesse Lumarie · [05:16]

指向原始笔记的链接

所以他们打算跳出窗户，然后拿着霰弹枪从前门回来。唯一的问题是，他们猜错了公寓。
And so they were going to jump out of the window and come back in through the front door with a shotgun. The only problem was that they had guessed the wrong apartment.
—— 嘉宾 · [10:14]

指向原始笔记的链接

从那以后,那套东西几乎全被扔出了窗外,现在流行的是提示词工程、上下文工程、循环工程。
Almost all of that has gone out the window since prompt engineering and context engineering, loop engineering.
—— Robert Lucero · [25:19]

指向原始笔记的链接

如果你能影响或在那些 token 里留下印记，那价值比出现在单个上下文窗口里高出数量级。
if you can get influence or imprint in those tokens, it’s worth, like, orders of magnitude more than being in a in a in single context window.
—— Ben Moore · [52:00]

指向原始笔记的链接

模型是 CPU,上下文窗口是内存(RAM),而记忆是磁盘。
The model is the CPU, the context window is the RAM, and the memory is the disk.
—— Dylan Couzon · [06:46]

指向原始笔记的链接

无限的上下文窗口是一个神话。
The infinite content window is a myth.
—— Elizabeth Fuentes Leone · [00:34]

指向原始笔记的链接

有时候内容窗口在毒害智能体，因为你的内容窗口里可能有提示词注入。
Sometimes the content window is poisoning the agent because you can have a prominent injection inside your content window.
—— Elizabeth Fuentes Leone · [02:54]

指向原始笔记的链接

在智能体之间共享上下文窗口，就像是给其他智能体添加垃圾，因为并不是所有智能体都需要彼此之间的信息。
You share the context window between agents is like add garbage to the other agents because no all the agents need the information between them.
—— Elizabeth Fuentes Leone · [09:59]

指向原始笔记的链接

当每个人都想要个人智能体，或者每个企业都想要自己的、全天候 24-7 持续在线并运行大量自动化的个人智能体时，上下文窗口比内存中的模型权重更重要。
When everybody wants personal agents or every business wants its own personal agents that’s always on and working 24-7 and running a lot of automations, the context window matters even more than the model weights in the memory.
—— Arvind Srinivas · [30:26]

指向原始笔记的链接

② 出现在这些集

1 集

• 《把训练好的模型当数据喂给模型：权重空间学习》 — 作为概念

③ 关联

点进去有真内容 —— 本页主要出口

Sam Charrington · Damian Borth · Hugging Face · 权重空间学习 · 权重 · 自编码器 · 模糊权重 · 分词 · 数据集提示 · 神经架构搜索