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

Qwen 3.8 27B is excellent, but it defaults to overthinking things

simonwillison.net

Post reply on HN
Print view View original

381–390 of 411 posts

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#381

Post by zahlman »

akie wrote: post #223
zahlman wrote: post #215

Earlier quoted context omitted.

At the bottom of this very submission are a bunch of dead comments that are very obviously LLM-generated.

Ok, you and I can easily spot LLM text. So what? The Turing test has still been passed, as is clear by people falling in love with ChatGPT, not believing something is AI, and by continuously claiming this or that is a bot. People, many of them at least, cannot make this distinction anymore. You can, I can, but people as a whole are having problems with that.

By this standard, the Turing test was also passed by ELIZA, but nobody serious actually gave it that credit. Aside from which, the understanding of that concept in popular media (both the nature of the test itself, and its supposed significance) has drifted way away from what Turing was saying.

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#382

Post by zahlman »

CamperBob2 wrote: post #353
zahlman wrote: post #210

Earlier quoted context omitted.

> Not if an LLM over chat can fool most people they're talking to a human (which it can) I keep hearing this claim, and yet I keep seeing LLM output which is trivially distinguished from human writing. I really can't understand how this gap persists; but then, there seem to have been at least some people who couldn't sniff out ELIZA, back in the day, too.

I keep hearing this claim, and yet I keep seeing LLM output which is trivially distinguished from human writing. That's because they're trained that way. If you trained a modern frontier LLM with the explicit goal of passing the Turing test, it would have no difficulty doing so.

Turn on showdead and look at the killed comments on this thread. (Every thread remotely related to LLMs seems to attract this behaviour.) If it's so easy to get it right, how is there such a high fraction of failures?

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#383

Post by CamperBob2 »

zahlman wrote: post #382
CamperBob2 wrote: post #353

Earlier quoted context omitted.

I keep hearing this claim, and yet I keep seeing LLM output which is trivially distinguished from human writing. That's because they're trained that way. If you trained a modern frontier LLM with the explicit goal of passing the Turing test, it would have no difficulty doing so.

Turn on showdead and look at the killed comments on this thread. (Every thread remotely related to LLMs seems to attract this behaviour.) If it's so easy to get it right, how is there such a high fraction of failures?

You've hit on the load-bearing crux of the matter. I have no idea.

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#386

Post by HDBaseT »

apples_oranges wrote: post #277
syhol wrote: post #221

Earlier quoted context omitted.

The word "open" has lost all meaning

Yes but investors and users seem to react well to it. Opencode for example gets mentioned more often than Cline even though it’s arguably inferior, openclaw is another example. Just name your stuff openX to win..

OpenCode is decent. I'm unfamiliar with Cline, what does the Cline CLI do better than OpenCode?

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#387

Post by pdyc »

madduci wrote: post #263
pdyc wrote: post #122

Earlier quoted context omitted.

i have same 255h and i was able to run it with low token speed 6-8tg/s with approx similar context window 60k

Interesting. What are you using? I was using ollama

i am using llama.cpp with pi harness on linux(cachyos). some of the things like coopmat are disabled in linux due to intel driver issues otherwise performance would be even better.

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#389

Post by scotty79 »

SwellJoe wrote: post #4

This is true, but I think it understates the problem. I did a task I've done with a bunch of small models lately ( https://github.com/swelljoe/flar/pull/17 ), and it did an excellent job, the best of any self-hostable model. But, it took eleven (11!) hours on my dual GPU setup. It really chewed on it, and spent a lot of time checking and re-checking. It is by far the slowest model I've used for the task. GPT 5.5 did…

> But, it took eleven (11!) hours > ... > most small models needed a couple of hours (but did a worse job)

I think a factor of 2 or 3 is a small price to pay for the best result.

Re: Qwen 3.8 27B is excellent, but it defaults to overthinking things

#390

Post by qeternity »

janalsncm wrote: post #374
qeternity wrote: post #272

Earlier quoted context omitted.

Jevons paradox: large purpose-fit data centers increase efficiency such that you can use AI in more places, and use more tokens for those tasks. The future is not a single chat bot session of bs=1. The future is many agents performing many tasks in parallel for a single user. Large GPU clusters will always have the edge in efficiency.

Jevon’s paradox suggests that total datacenter resource consumption will increase. It doesn’t say that people will choose datacenters over their own personal hardware when the latter is sufficient.

Personal hardware is only sufficient today for some tasks. As data center power efficiency, and large sparse MoE task efficiency increase, personal computing will continue to lose out.
Post reply on HN