[TylerE]

No, it’s just a fundamentally much better model. Going back to Opus feels like the model has been lobotomized. It makes much more frequent errors, especially of the “I claimed I tested x y and z, but actually only kinda half heartedly tested x, and assumed I understood what was wrong” variety.

[hypfer]

Wait but that has been the exact word-for-word complaint when comparing sonnet to opus

Or opus to opus

Or really any new thing to old thing

[solumunus]

When the agent is becoming more accurate and thorough what would you expect to be reported?

[hypfer]

Oh I am sure that it became somewhat more accurate, and with that, the labeling there is in fact technically correct. It just does not work as an explainer for the doomsday-ish hype that model has induced in a lot of people's brains.

The user here is right in what they said but wrong in why they said it, essentially.

[ben_w]

An analogy I keep coming back to with the current progress in LLMs is the progress in the 90s of 3D game engines.

Every upgrade made what came before it appear awful in comparison, to such an extent that every upgrade was called "photorealistic" and people kept forgetting that they'd been using that description for the previous engines that they were now dismissing.

https://archive.org/details/nextgen-issue-26

[TylerE]

That’s a rather bad faith framing, I think. Who are you to judge why I said something?

[hypfer]

A person with the exact kind of pattern matching brain disorder this tech has been modeled after.

I do make mistakes though. Please check results.