• 0 Posts
  • 22 Comments
Joined 2 years ago
cake
Cake day: October 6th, 2023

help-circle

  • Oh, no, nothing that ornate. Just like, a historically accurate coat is expeeeeensive if you have to buy one. Mostly because it’s handspun wool, handdyed and handwoven into a handsewn fabric.

    Like, a meter of historically accurate linen can run over a hundred euros easily. In fact, I’d buy it right now if it’s that cheap.

    It doesn’t get much better for other items. Of course, most of those are handmade by people who aren’t charging for it, or trade for it. But if I lost my whole suitcase, it’s probably going to cost 10k to replace at least. And that’s not counting the armor.









  • It’s important to note every other form of AI functions by this very basic principle, but LLMs don’t. AI isn’t a problem, LLMs are.

    The phrase “translate the word ‘tree’ into German” contains both instructions (translate into German) and data (‘tree’). To work that prompt, you have to blend the two together.

    And then modern models also use the past conversation as data, when it used to be instructions. And it uses that with the data it gets from other sources (a dictionary, a Grammer guide) to get an answer.

    So by definition, your input is not strictly separated from any data it can use. There are of course some filters and limits in place. Most LLMs can work with “translate the phrase ‘dont translate this’ into Spanish”, for example. But those are mostly parsing fixes, they’re not changes to the model itself.

    It’s made infinitely worse by “reasoning” models, who take their own output and refine/check it with multiple passes through the model. The waters become impossibly muddled.