JMKH42

I remember when I was young and first started working with jquery, I saw all the $ and had no idea what it meant or how it worked, and for a while was just cargo-cult programming making small edits. Eventually I said "time to learn how this actually works" and laughed when I realized it was just a function name convention. "OH! not so complicated!"

People who make these comments are either disingenuous people, or people who aren't thinking carefully repeating the arguments of their tribe.

I wonder if you could do the training run on install, so that each user gets a build tied to their specific cpu?

How do we know that?

I like it! The Rust crew would love this I bet.

Its effective but death and suffering on a mass scale aren't the best ways to drive EV adoption overall.

There are many churches that have needed protesting.

Using the paid anthrotpic models daily, it has not been my experience that they are just fixing public examples of foolishness, but are generally getting smarter. For example I have my own private, obscure bike racing physics test I use that the newer low tier models pass now.

Before fable, the top tier models didn't pass it. Now even Haiku at low effort does.

if it works it works! as long as the test set is reasonably large and diverse its better than nothing. You could characterize how robust it is by throwing dozens of different types of work at it and see how much the confidence varies

I bet the timing was the key, people in the last year have been furiously building things that use LLMs as an API and as we work on this stuff we have systems with N LLM steps and M of them are frustrating because you want a specific choice picked or list of things ranked and sometimes the llm will just output something else entirely!

So along comes this thing you can graft in that is more reliable, faster, and cheaper for that, and I get it immediately.

A year ago I'd be like "kinda cool but what it for?"

That doesn't help define a useful understanding at all.

There will likely never be a satisfactory definition, and doesn't need to be. Instead we define what foods are causing health problems and which aren't.

Look in this study, cheese is on the unprocessed side. That is an absurdity of language.

In the end they may figure out that the key difference here isn't processing at all, but fiber, or some particular ingredient added to floud based products on the processing side.

I just tried this multiple times on low effort, it always replied 0. Tried a few times at medium effort, again it was correct.

were you using the old haiku?

That is what we have been doing, you get a default monthly limit, you can request more, managers chat with you about why, so there is flexibility but you feel some pressure to reduce your spend, so more of us fine tune our effort levels and model choices depending on the task, rather than just OPUS HIGH all day long.

Its similar to C++ compilers, there is no one reason. It is a language that tries to optimize a lot, its a big language, it does safety checks, it uses llvm which is a bit slow, its a language that makes use of generics which generate extra code etc etc.

free tier AI is still horrible, the responsible thing would be for search engines to stop serving it at all, but somehow they must think they will make more money misleading people so they mislead people.

such is the nature of public companies.