Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claudette: Make Claude stop talking like a BuzzFeed article

github.com

Post reply on HN
Print view View original

241–250 of 254 posts

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#241

Post by quibono »

Similar to vomit[0] and claudish-to-english[1].

My general feeling here is that this really needs to be fixed at the model level. All other solutions are mere hacks that work for a bit but all have drawbacks. It's either adding yet more context (which has variable rates of success) or piping to a different model to translate, or using a skill, or mentioning this per each prompt, or....

At this point I think it's in Opus's nature to talk like this, and given enough back-and-forth it will start to drift back to its classic mode, aka this incomprehensible "load-bearing seam" babble.

[0] https://github.com/zachahn/vomit/tree/main

[1] https://github.com/gvzdv/claudish-to-english

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#242

Post by ianjbutler »

zarzavat wrote: post #191
ianjbutler wrote: post #148

Earlier quoted context omitted.

I hear that, sort of, but here's the thing. Using AI at scale means AI needs to be nearly perfect about not shitting where they eat. That's the subject matter of the whole thread So the options are a) being a really aggressive stickler for generative hygiene with deterministic rules, b) being massively wasteful about hiring a few machine janitors for every machine coder, or c) humans become the machine's janitor. If…

"AI at scale" is just a euphemism for slop. Current gen AI can augment human engineers but not outright replace them.

This isn't really responsive to what I'm saying or what the discussion here is about, but if you insist. Would you describe lots of AI augmenting lots of human engineers as perhaps.. AI at scale?

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#243

Post by suika »

The sort of recent increase in these "claudisms" finally made me buy a ChatGPT Pro subscription. First time using Codex CLI. It's surprisingly different, feels good/fresh. I guess one can say I was quite loyal to Claude Code, or maybe just lazy / it worked well enough so far. And now I'm considering finally cancelling my Claude Max subscription. It would be another first time too.

It was mentioned that ChatGPT (still) has some similar quirks too. What are they? Maybe I actually shouldn't look out for them, so that the freshness of ChatGPT will not be lost.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#244

Post by YuriNiyazov »

rnvannatta wrote: post #238
YuriNiyazov wrote: post #68

Earlier quoted context omitted.

It's easiest to explain this while anthropomorphizing the model, I know some folks here hate that, sorry about that. I heard an interesting diagnosis for why Claude does this: the output is a compressed version of its thought traces, very dense because the model is under pressure to use as few tokens as it can and to pack as much (for accuracy) of its concepts into the output. One of the reasons that "don't do X" typ…

In addition to positive reinforcement on the style I wanted for output, I tried your suggestion to tell Claude Fable to think Claudese in its thoughts, and that sadly got the session immediately flagged as too risky for Fable to handle, suggesting I downgrade to Opus. Several tweaks still consistently flagged it until I removed the directive about its thinking. So I guess Anthropic wants zero mention of thoughts with…

Very interesting, thank you!

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#246

Post by cheikhcheikh »

enobrev wrote: post #231
bhelx wrote: post #114

Earlier quoted context omitted.

This speaks to the general problem with using LLMs for writing. The audience they are writing for us you, but you're trying to write for a totally different audience. In code, this manifests as comments in the code that are hyperspecific to the conversation you are having, and not the long term benefit of having those comments in the code. I see this in docs a lot. I've been reading a lot of docs these days where it…

It reminds me of something I've always wanted as a coder but never cared enough to implement, which would be a verbosity switch. I see the full multi-paragraph comments in my codebases and get annoyed but also feel like the additional context helps improve the llm results over time because that history helps it know what's been tried and removed in the past. It's additional context for the system that improves with c…

What you want can be accomplished with an extra doc, call it the Log, where the llm appends things tried, lessons learned, failed experiments etc.. while leaving the comments as terse accurate snapshot of the current state. I've been using this log pattern and it works well.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#249

Post by klardotsh »

hilariously wrote: post #205
klardotsh wrote: post #175

Earlier quoted context omitted.

This one drives me nuts, especially because I divvy up plan documents into such granular sub chunks and Ralph loop over them, I get nonsense commit messages and comments like “PLAN-5.1.A.d.42 load bearing reassertion” unless I AGENTS.md to hell and back… and still end up having to manually reject 5-10% of commit messages because the agent simply forgets AGENTS.md instructions until reprompted.

Sometimes you can do a separate skill for committing only and basically minimize the context - "analyze the commit messages, the original instructions, and this set of things to stop wasting my FUCKING TIME"

I've definitely considered writing a commit skill and a `cc-safety-net` custom rule to fully forbid `git commit` other than with `--amend` ++ `--no-edit` outside of that skill. Hm, maybe that's a task for an upcoming weekend.

Re: Claudette: Make Claude stop talking like a BuzzFeed article

#250

Post by ipdashc »

lloydatkinson wrote: post #194
ipdashc wrote: post #192

Earlier quoted context omitted.

I'd seen it around but assumed it was an effect of how you prompt it or something. And figured people were overexaggerating a bit when they complained about it. Nope, I just tried it out myself recently and... wow. In the very first conversation it started glazing me about being right to push back, having the crucial insight, and something something the load-bearing-whatever. So yeah, I'm on the same page as you, how…

What system prompt did you use?

I more or less just wrote "never say stuff like 'good insight' or 'you're right to push back', and avoid addressing me directly, just answer questions". I'm sure there's better ways to write it, but meh, it seems to have worked well enough
Post reply on HN