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

GPT-6 Astra, looped transformers, and hidden reasoning

magazine.sebastianraschka.com

Post reply on HN
Print view View original

131–140 of 153 posts

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#131

Post by OneOffAsk »

cainxinth wrote: post #25
siva7 wrote: post #18

Astra was insane until Monday but something happened on tuesday, now it feels like Sol. I grieve for the lost productivity but i hope they may give us the original Astra back.

It's the same story every time OpenAI or Anthropic releases a new model. They are generous with compute for the first few days, and use maximum fidelity with uncompressed weights. Everything runs at its best to make a good first impression. But eventually they pare things back and the models perform a little worse.

It’s all speculation (you too), but I think the effect you’re describing is instead getting calibrated to the model’s limits. Next time a new model comes out, wait a month before trying and see if you have the same feeling of rapid quality decline after a few days. I did after I jumped back into it mid 5.x or whatever ChatGPT after paternity leave. Blown away for a few days, worried about my job for a few days, then increasingly aware of its limits.

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#132

Post by lann »

olalonde wrote: post #130
silversmith wrote: post #65

Earlier quoted context omitted.

Why is your agent able to call ssh. Why can it trigger 1password. Why are you giving metaphorical guns to metaphorical toddlers. Why is it not sandboxed. Your practices worry me.

Are you guys all running agents in VMs?

Yes.

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#133

Post by enraged_camel »

AnimalMuppet wrote: post #73
enraged_camel wrote: post #24

Earlier quoted context omitted.

>> It's over zealous at times (which is why I stopped using Claude) and gets too creative when doing agentic system level stuff. Accessing files and doing things it shouldn't do. I gave Astra a pretty straightforward bug ticket yesterday. The bug involved an edge case that could sometimes result in an invalid value getting stored in a user profile field. Pretty harmless, no crash or anything, just annoying. Based on…

You have to watch it like a hawk so it doesn't do something to production, on its own, without a specific request? Wow. Then I could never trust it to not be doing something to some other system that it shouldn't, so I'd have to audit every network request. If enraged_camel had been doing something else involving the production database at the wrong time, they might have accepted the 1Password prompt.

Worth noting that this has never, ever happened with Anthropic models, which I've been using all day every day since Opus 4.1.

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#134

Post by kadoban »

dooglius wrote: post #111
simlevesque wrote: post #78

Earlier quoted context omitted.

We can't have hard evidence. It's a SaaS and they own the code and the machine it runs on. So it may be a widespread hallucination. But there's no evidence of that either.

Run a benchmark with a large number of samples, rerun a few days later. Compare results, use statistics to see if there's a statistically significant difference.

Doing this in a way that doesn't get you noticed or fucked with is potentially going to be quite difficult.

If they have a "hey we're being benchmarked" mode, which is not hard to imagine, avoiding tripping it is going to be annoying and difficult to prove.

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#136

Post by majormajor »

Vetch wrote: post #80
cainxinth wrote: post #25

Earlier quoted context omitted.

It's the same story every time OpenAI or Anthropic releases a new model. They are generous with compute for the first few days, and use maximum fidelity with uncompressed weights. Everything runs at its best to make a good first impression. But eventually they pare things back and the models perform a little worse.

The most charitable explanation I can think of for this is something like regression to the mean. When a model is first released, there'll be a subset of users who, just by chance, sample the highest quality band of the distribution that answers their query. Some of them will rush over to social media and post about how amazing a model is. Over time, those users' mental model of responses will converge but they'll pe…

> I also think there's an aspect where initial testing is more forgiving because the more persnickety polish bits can be ignored and tests are likely to have similar structure to things that can be trained for. Meanwhile, actual specific work items are a broader unusual distribution with more stringent acceptance criteria.

IMO this is 90% of it (as someone who has a bit of a different interaction style and runs these things less autonomously, and hasn't generally seen the claimed regressions). Day 1: throw new stuff at it that failed badly, exciting to see something make more progress! Day n: reality sets in that it still wasn't perfect the first time.

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#137

Post by majormajor »

kmeisthax wrote: post #105
Vetch wrote: post #80

Earlier quoted context omitted.

The most charitable explanation I can think of for this is something like regression to the mean. When a model is first released, there'll be a subset of users who, just by chance, sample the highest quality band of the distribution that answers their query. Some of them will rush over to social media and post about how amazing a model is. Over time, those users' mental model of responses will converge but they'll pe…

To add onto this, if you use a shiny new model and it gives you a turd, you're not going to tweet about it ("hey guys, look what I made with Astra! Nothing!"), and even if you do nobody is going to interact with it so it does poorly in the algorithm, because it has to compete with all the people using the new model to make something that looks impressive. Then people get tired of the magic trick and the logic flips.

We've seen that some---gpt5 was considered pretty lackluster intially, in particular. Opus 4.7 and 5 vs 4.6 were also greeted with a lot more "meh" than 4.6 or Fable.

Re: GPT-6 Astra, looped transformers, and hidden reasoning

#139

Post by tesnorindian »

I gave owao/Nanbeige4.2-3B-GGUF (Q8 quant) a try to understand how loop transformers work and compare it with other models especially with Ling 3 Tiny MoE model. As reported in the article, it is compute intensive (due to looped layers) and made a mistake during tool call just like how Ling 3 Tiny MoE did for exactly the same prompt.
Post reply on HN