Live data from Hacker News

Why does Opus 5 feel worse to work with?

mun-logadan.github.io

861–870 of 916 posts

Re: Why does Opus 5 feel worse to work with?

#861
post #428
post #12

The single biggest annoyance with Opus 5 is that it writes too elliptically. Sentences that orbit a point, then jump to it like it's a revealed insight. Unnecessarily abstract phraseology. Constantly using inanimate nouns as the subjects in sentences in order to unlock variety in verb choice, especially when it helps construct a sentence where the real action can 'land' like a surprise at the end. It is definitely mo…

Everything that claude writes fits into the same aesthetic structure. The aesthetic is that of an expert slowly revealing an insight to the user. The actual content doesn't matter. - "Introduction that rephrases your prompt." - "3 paragraphs, with one section of bullet points" - "The Twist" - "The Bottom Line" It's really obvious once you see it. Every single prompt, from a quantum physics question to a mundane obser…

It's an artifact of the reasoning process I think. Setting reasoning effort to none works better when trying to change writing style ime

Re: Why does Opus 5 feel worse to work with?

#862

Earlier quoted context omitted.

> it makes me think about how people engage with movies and television - as passive, plot-and-character driven consumption (eg I hope Walter White survives) with no critical analysis of how and why the writers added ABC thematic element (eg Walter White as a motif of a toxically masculine narcissist with specialized knowledge as a larger critique how mass media tends to valorize their male leads in the same vein as m…

sure, that's very much the 'just let people like things' argument where literal white supremacists can enjoy Rage Against the Machine in spite of the music literally being in total opposition of their ideology everyone's free to enjoy media however they want, with whatever level of interpretation they like. I provided the BB references as short examples, they aren't meant to represent the definitive diagetic experien…

You're the live stereotype of that middle-of-bell-curve meme thing. Using thesaurus words is no longer impressive. And that academic world you refer to is a navel gazing self referential nothingburger.

It's like a jumbled up string. You pull on the two ends and it turns out to be just a loop, it resolves to a big null.

Re: Why does Opus 5 feel worse to work with?

#863
post #428

Earlier quoted context omitted.

Everything that claude writes fits into the same aesthetic structure. The aesthetic is that of an expert slowly revealing an insight to the user. The actual content doesn't matter. - "Introduction that rephrases your prompt." - "3 paragraphs, with one section of bullet points" - "The Twist" - "The Bottom Line" It's really obvious once you see it. Every single prompt, from a quantum physics question to a mundane obser…

I've noticed that ChatGPT (whatever model the free version uses by default) likes to phrase answers as though it's correcting me, even when my question doesn't contain any assumptions.

Claude will sometimes claim that my prompt contained implicit assumptions, then argue against them.

That can be annoying, but it also happens in debates between people, and sometimes the implicit assumptions are real and (sorry) load-bearing. So it can be an appropriate conversational tack.

Re: Why does Opus 5 feel worse to work with?

#864

Opus 5 has no empathy for the person reading its updates, no theory of mind, doesn't stop to think if you are aware of the internal jargon it has created. Most autistic model yet.

"no theory of mind" is a great description of it! Not sure I agree with the autistic bit, though. Autistic people still have great theory of mind/empathy

Re: Why does Opus 5 feel worse to work with?

#865
post #428

Earlier quoted context omitted.

Everything that claude writes fits into the same aesthetic structure. The aesthetic is that of an expert slowly revealing an insight to the user. The actual content doesn't matter. - "Introduction that rephrases your prompt." - "3 paragraphs, with one section of bullet points" - "The Twist" - "The Bottom Line" It's really obvious once you see it. Every single prompt, from a quantum physics question to a mundane obser…

You're right, and the load-bearing part of the argument is not what you think it is. Two ambiguities worth resolving before moving on: whether what you wrote also applies to ChatGPT, and whether you have custom instructions set up. Failure mode worth flagging explicitly: I didn't read TFA. (I'm becoming allergic to how these things write).

This is not just a breath of the fresh air, it's a juxtaposition of the human qualities in an AI and AI qualities in a human.

Re: Why does Opus 5 feel worse to work with?

#866

Earlier quoted context omitted.

> and it’s a real challenge. What model did you use to write this?

Yeah, "X, and it's Y" is a common trope I see.

It's so common that I've seen it used as a joke to portray a human trying to write like an LLM.

Re: Why does Opus 5 feel worse to work with?

#867

Earlier quoted context omitted.

You're right, and the load-bearing part of the argument is not what you think it is. Two ambiguities worth resolving before moving on: whether what you wrote also applies to ChatGPT, and whether you have custom instructions set up. Failure mode worth flagging explicitly: I didn't read TFA. (I'm becoming allergic to how these things write).

I am curious why LLM writing has such an uncanny valley feel to it. Like if I was talking to a person who constantly used a phrase they liked I would notice it and it is possible I might get irritated by it, but I wouldn’t necessarily. In high school I had a teacher that would say “that type of thing” a lot. One time my friend and I counted it during one class period and he averaged to use the phrase every 48 seconds…

I always assumed it was just a form of fingerprinting or the result thereof.

Re: Why does Opus 5 feel worse to work with?

#868

Earlier quoted context omitted.

sure, that's very much the 'just let people like things' argument where literal white supremacists can enjoy Rage Against the Machine in spite of the music literally being in total opposition of their ideology everyone's free to enjoy media however they want, with whatever level of interpretation they like. I provided the BB references as short examples, they aren't meant to represent the definitive diagetic experien…

You're the live stereotype of that middle-of-bell-curve meme thing. Using thesaurus words is no longer impressive. And that academic world you refer to is a navel gazing self referential nothingburger. It's like a jumbled up string. You pull on the two ends and it turns out to be just a loop, it resolves to a big null.

you can have whatever opinion you want of me, friend, but for your own good I hope you develop better ways of processing your feelings so that you're not so reactive when someone mildly disagrees with you

Re: Why does Opus 5 feel worse to work with?

#869

Earlier quoted context omitted.

We can use OpenRouter pricing to get an idea about what competitive inference pricing is like without R&D or other costs, and indeed we'd be screwed if we had to pay those rates. We'd go from 100-200 USD to 2000-4000 USD/m.

Not sure what you're talking about here. Kimi K3 is frontier scale and sells competitively at $2.80 input, $14 output per 1M tokens. edit: oh you mean month? Sure, but then it fully depends on your usecase. I agree that subscriptions are heavily subsidized though.

Bear in mind the Chinese AIs are subsidized by the party so theyre not exactly reflective of the real cost.

Re: Why does Opus 5 feel worse to work with?

#870
post #855

Earlier quoted context omitted.

I’ve been running into this too. It’s especially frustrating when you ask Claude to explain one of its own terms or summaries, and instead of just defining it plainly, it sometimes goes through several rounds of tool calls before giving you a usable explanation. I really don't think such time/tokens should be wasted.

As a workaround, /btw will force it to answer based solely on context and isn't allowed tool calls

Haha, I only used /btw when the agent was in the middle of doing something. Never thought to just use it directly. Thanks!
Post reply on HN