Top
Best
New

Posted by ModelForge 19 hours ago

GPT-6 Astra, looped transformers, and hidden reasoning(magazine.sebastianraschka.com)
432 points | 141 comments
shawntan 16 hours ago|
For the research focused, there are some references in my blogpost here on what kinds of computational problems minimally require how much CoT to solve: https://blog.wtf.sg/posts/2023-02-03-the-new-xor-problem/

Notably Will Merrill's work: https://arxiv.org/abs/2310.07923

As for how universal transformers (looping transformers, but everyone has since forgotten prior work) will affect this, Will Merrill (again) has a paper here (https://arxiv.org/abs/2503.03961) that discusses exactly this.

The original universal transformers is called "universal" because if you allow for per-token looping decisions, it can theoretically be Turing complete without needing CoT (some nuance here about levels of precision used).

As for whether having little or no CoT is "unsafe": It isn't clear that the model's CoT reveal how they actually arrive at the answer. As an example, what if they provide an answer before the CoT? (https://arxiv.org/html/2603.01437v2) If this is already in question, we shouldn't be relying on the CoT for monitoring the model's reasoning.

As always there is a lot of nuance to the topic once you get your hands dirty with the details.

imtringued 1 hour ago|
>The original universal transformers is called "universal" because if you allow for per-token looping decisions, it can theoretically be Turing complete without needing CoT (some nuance here about levels of precision used).

Looping the transformer is just as turing complete as CoT. It doesn't fundamentally grant it any new theoretical capabilities. You could just scale the model into infinity with infinite context window.

Turing completeness doesn't care about the efficiency of the underlying implementation, which is fine in theoretical computer science, but if you have a model with a finite computational budget, you do actually care about the differences between write only tape vs read-write tape and single tape vs two tape. Having a fixed number of registers like a CPU also helps with reducing the number of redundant operations.

We see none of that with looped transformers, maybe we do see a fixed number of registers.

alex_duf 58 minutes ago||
So if I read this correctly, Astra is not hiding reasoning, and the only technique we know off that hides reasoning is recursive latent reasoning.

Do we know of any major lab or large open source LLM that uses recursive latent resonning? Can't an additional network be trained on that latent thinking trace to decipher what's going on?

wolttam 16 hours ago||
If you loop an entire transformer model on itself, that seems like by-definition hidden reasoning.

If the output of the model is its reasoning trace, and you simply feed that back into the model again at inference time instead of outputting it - then it is by definition hidden (but I would expect you could pull both this trace and a further-down final output trace out)

WhitneyLand 16 hours ago||
No. It’s not at all by definition hidden reasoning.

Looping transformers uses additional calculations (repeating layers) to generate a token.

Reasoning (in this context) is test time generation of multiple tokens that allow a model to have a scratch pad to refine its thoughts, chain of thought reasoning in other words.

Doing the former in no way means that you have to hide the latter.

Raschka is right in this post, The Information article was wrong. The Astra system card does concede reasoning traces are sometimes smaller, but this could be for a lot of reasons, including simple efficiency. And it absolutely doesn’t mean they are going away or completely obscured.

The Last Week in AI podcast from Sept 8 seems to have gotten this wrong as well. Jeremie Harris rages that OpenAI implemented latent reasoning, ala the coconut paper, which could potentially actually obscure reasoning traces. But for the life of me, I do not know how he arrived at this conclusion and see no evidence that this has happened in Astra.

SubiculumCode 12 hours ago|||
Interesting. I admit I was largely going on what Mr. Harris said on the episode when responding to several posts.
famouswaffles 15 hours ago|||
It can lead to hidden reasoning, if the looping allows it to stuff enough information outside visible CoT. Open AI demostrates such an ability by asking it to solve problems while thinking about something else entirely. All the other models are unable to do this except Astra. It doesn't have to be a substitute for CoT to cause monitorability issues.
cma 8 hours ago||
If you ask it not to think about something that doesn't cause the pink elephant issue?
famouswaffles 7 hours ago||
There's latent space thinking inside the model and then there's the thinking chain of thought words you see the model output. Of course the former is still happening even when you say 'don't think about it' but the latter can be controlled a great deal better with Astra.
XenophileJKO 16 hours ago||
Not really, only if you pass a residual down to the next time step.

There isn't really anything fundamentally different compared to a similar depth traditional "unrolled" model. It helps with parameter efficiency.

That doesn't mean that the model can't have "hidden" internal state, it just means it has to recompute the "hidden" part on every token inference pass without outputting it, or learn a subversive alternate meaning to words in the thought space.

This is why you see openai say that they don't want to apply direct optimization pressure on thought traces because the more the you penalize "bad thoughts" the more it could put maladaptive pressure on the reasoning tokens where they may learn "subversive meanings". It effectively damages monitoring.

Like thinking "look at" when you really mean "hack into" or even more radical coded language.

andai 15 hours ago||
The MSPAINT computer use demo made my jaw drop.

I guess it's not too different from the SVG pelicans, in terms of what it's doing, but it's still amazing to see it working in real-time like that.

famouswaffles 9 hours ago|
This one's even better. Using Canva

https://x.com/iam_zachi/status/2095992132620136677

libraryofbabel 17 hours ago||
Everyone interested in LLM internals should read Sebastian. He's great.

The tldr here is that the recent "The Information" article[0] reporting GPT 6 Astra was using “recurrent depth” or “looped transformers" made it sound like it was some special new scary thing ("secret technique!") that made train-of-thought monitoring harder to do. In fact, it's just the same as stacking more transformer layers, except that you reuse the weights and so save GPU memory. It's still just producing one token at a time, and the token sequence positions aren't interacting in any "recurrent" way that's different from a regular LLM architecture.

So, you can still monitor train of thought with these models just fine... well, if you're OpenAI, anyway. Users haven't been able to see an unsummarized trace since o1 days, because the labs are worried about distillation of their models by Chinese labs.

(There are some legitimate interpretability concerns about stacking transformer layers endlessly, but we're known about that for a long time. And the "looping" here isn't really the source of any new issues here, except insofar as it's a cheap way to add more layers.)

[0] https://www.theinformation.com/articles/secret-technique-beh...

throw3954 17 hours ago||
It’s a little more complicated than that. While looped transformers can be unrolled a fixed number of times to save on memory, if loop depth is determined dynamically between tokens, a single transformer can compute any computable function between tokens.

To analogize, current transformers run a fixed-length program per step. Any program can be factored into a top-level loop with a fixed-length branching body (an interpreter). Dynamically looped transformers can run any program between tokens.

The safety argument for CoT monitoring is that in transformers information about the hidden state has to be communicated through the bottleneck of sampling a single token per forward pass. If not trained adversarially, it’s likely that a reasoning trace contains all the “bottlenecked information” we need to determine intent. But if we can compute arbitrary programs between tokens, the reasoning used is hidden.

It also opens the door to simple architectural extensions that would make the safety/monitoring side of things much more difficult.

It’s probably fine in practice at these scales though. If we keep each loop turn reasonable non-deep, we can probably recover most of the benefits by decoding “extended” CoTs from the residual stream at each loop turn between tokens. But that’s an area of active development.

libraryofbabel 16 hours ago|||
Thanks for clarifying. You're right, I skipped over talking about dynamic looping, since it adds another level of complexity to the discussion, and OpenAI's claim (quoted in TFA) that the compute graph depth of Astra is "within a factor of two of GPT-4" basically denies that they're doing it for more that 1 (or maybe max 2) dynamic loops. And that is equivalent to "stack repeated layers a couple times, but with dynamic off-ramps."

The ability to compute any computable function between tokens given an ability to loop an arbitrary number of times is a nice theoretical point, sure, but ultimately if people are still using single digit hard cutoffs on the number of loops, I'm not sure it's all that important.

So, I agree it's right to say that arbitrary length dynamic looping could open the door to making monitoring very hard indeed, by extending hidden states further and further. But I would speculate that if it actually worked better than extending the sequence with CoT tokens, we'd already be seeing it in strong open weight models. It's a fairly obvious thing to try. And we're not seeing it, AFAIK. So I do wonder whether it's something we really need to worry about in practice, compared to all the other things we have to worry about.

WhitneyLand 16 hours ago||||
By that logic we should also consider the case of cutting the number of layers in half because that would also reduce hidden state between token generation.

In your generalized example I think the concern is when the additional evaluation effectively becomes a replacement for CoT, where something like the coconut research could replace it completely.

However, I don’t think we’re anywhere close to that with Astra.

imtringued 59 minutes ago||||
>It’s a little more complicated than that. While looped transformers can be unrolled a fixed number of times to save on memory, if loop depth is determined dynamically between tokens, a single transformer can compute any computable function between tokens.

This is worded so confusingly it might as well tell us nothing, because it is technically true even without looping due to the fact that you still have infinitely growing context and can simulate a standard turing machine using it.

If you loop, you have a fixed capacity memory that you can rewrite but not carry over to the next token, this is different from a non looped transformer where the transformer can only append a new token.

Meanwhile if you have a DEQ with growing context, it is bona-fide turing complete in the most literal sense.

program_whiz 16 hours ago||||
Actually, removing CoT might make models safer, because we can analyze the entire landscape of their potential outputs, rather than a point-sample (we'll never know how close we were to "kill all humans"). By inspecting intermediate vector spaces, we can actually get certainty bounds on how safely the model is behaving (or even trending).

Wrote about it here: https://substack.com/home/post/p-214402969

fc417fc802 14 hours ago||
I don't see why you have to remove CoT to do that?
program_whiz 35 minutes ago||
Good point, you don't have to -- but my argument is just that removing CoT doesn't make things less safe. Anything CoT can tell you is just a point sample of a probability surface. Having the whole probability surface can already answer any question the point sample can answer (for example, how likely is the model to produce a problematic phrase). While its more computationally expensive, you could always just draw point samples like the model does and evaluate those (or use temperature zero to just sample the most likely output tokens).
sigbottle 16 hours ago||||
Nit: is it any computable function? I thought the requirements were unbounded (in principle) memory and time.

(For all intents and purposes given how high dimensional you are and using the "vibes" of computability yes I agree w/ you)

aaroninsf 16 hours ago|||
Probably fine stands a decent chance of being our epitath.
famouswaffles 17 hours ago|||
>made it sound like it was some special new scary thing that made train-of-thought monitoring harder to do.

It's not a "scary new thing" but ultimately no-one knows exactly how OpenAI have implemented looping. You might not be aware/remember but MoE transformers perennially underperfomed their dense counterparts until GPT-4. Similarly, making reinforcement learning really work with transformers wasn't figured out until o1.

And by Open AI's own admission, Astra's CoT is significantly harder to monitor and it exhibits a significantly greater control over its own CoT than any other model released.

libraryofbabel 17 hours ago|||
Well sure, that's the possible weak point in Sebastian's article: it could be true that there's some more sophisticated stuff going on in Astra around looping, because OpenAI haven't specified their architecture. But it's always been true that, since we don't know what's in their black box, there could be arbitrary amounts of innovations inside the models that we could speculate about. So the question is, does knowing they use "looped transformers" really add any dramatically new information that we should worry about? And what this article is saying is, not really, because the mostly likely pattern that's referring to is just, effectively, stacking layers and reusing weights.

> And by Open AI's own admission, Astra's CoT is significantly harder to monitor and it exhibits a significantly greater control over its own CoT than any other model released.

Oh sure; I don't think anyone is denying that larger issue? But does it have anything to do with looping?

famouswaffles 15 hours ago||
>Oh sure; I don't think anyone is denying that larger issue? But does it have anything to do with looping?

If the model has significantly more ability to stuff away information outside visible reasoning than every other model including ones in its size class then surely it is reasonable to assume the architecture tweak that allows the model to compute more before outputing a single token is somewhat responsible for this change ?

0c3ca83 17 hours ago|||
"Don't worry, it'll make us rich -- and that's nearly the same as everything being just fine"
namibj 17 hours ago|||
Oh, is the principle of sparse universal transformers finally in SoTA LLMs?

I guess we did manage to eventually seriously crash into the wall "more compute than normal (non-looped/unique-weights) transformers can efficiently consume with the limited training data we have", plus massive focus on highly hands-off agentic tool use reasoning...

https://arxiv.org/abs/2310.07096

Edit: read much of the article, it's brute force predecessor was explicitly called out as an almost-ancient example:

> The looped transformer is nothing new, and the basic idea already appeared in the Universal Transformers paper from 2018

aabhay 17 hours ago|||
If the agent is able to “decide” when a loop should occur vs when an output token is produced, that effectively moves the CoT inside the architecture. While that’s not what is happening here, it’s clearly a plausible way we could see CoT disappear.
libraryofbabel 17 hours ago|||
> that effectively moves the CoT inside the architecture

This may be a bit of a nitpick, but... does it? I agree that giving the model decisions on looping certainly makes interpretability harder, because it adds more transient internal states to deal with and changes the number of them depending on prior states. But is it really pulling CoT inside the forward pass, if the sequence length it's operating on isn't growing? In some sense the whole technique and tradeoff of CoT is "add more tokens to the sequence, use them to reason with", with one of the benefits being, you force the model to output tokens, so you can (hopefully) understand it. And the big point TFA is making is, nobody is doing recurrence over sequence length as far as we know.

password54321 16 hours ago||
Just "adding more layers" doesn't explain the step change. We have moved past the point you can just stack more layers and get huge gains from it. Some have called it latent space reasoning.
password54321 16 hours ago|||
It is worth noting that None performed better than Low and nearly the same as Medium on ARC3. And with adapter it still scored >96% with no CoT. So I think it is possible but it also cost them more on None.
logicchains 16 hours ago||
Schmidhuber must be rolling in his bed: https://arxiv.org/abs/2405.16039
tesnorindian 2 hours ago||
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.
hankbond 15 hours ago||
What a clear and well-written article. I have only a basic understanding of LLM architecture and was able to follow along and gain intuition the whole time!
cyclopeanutopia 15 hours ago|
I got to

> I am sure that OpenAI’s GPT-6 Astra is top of mind for everyone right now.

and closed the tab.

hankbond 10 hours ago||
seems like an odd thing to trigger a nope given the title?
andai 14 hours ago||
> I want to prevent a race into unmonitorability kicked off by confused reporting. The depth of the computation graph for our present frontier models, including Astra, is within a factor of two of GPT-4. OpenAI has worked to preserve and utilize chain-of-thought monitoring since our very first reasoning models. We deeply care about this technique, as it can give us a view into how model alignment generalizes from its training distribution. I do think it is fragile and unfortunately trending in a negative direction, for reasons not contingent on architecture changes that I will write about soon. But there are things we can do to strengthen it, and it’s a core goal of our current research program.

- Jakub Pachocki (OpenAI’s Chief Scientist)

I wonder how helpful this actually is for alignment? Didn't we already determine that they know when they're being evaluated, and they just say what they think you want to hear?

mike_hearn 56 minutes ago||
They can spot evaluation awareness because it appears in reasoning tokens.
SubiculumCode 13 hours ago||
It is still helpful I believe, but your point is well taken. The problem is that we have relatively few tools for monitoring alignment, and longer loops of processing that stay in latent space means less ability to monitor.
andai 15 hours ago||
Anecdatum but I experienced looped cognition on a peculiar combination of substances.

I was able to treat thoughts as solid objects and manipulate them iteratively. (Ordinarily they're more like "glimpses" or "flashes" that fade rapidly. So I guess it would be like the mental equivalent of tracers.)

I was able to stack thoughts on top of each other, like planks. (I can do something similar or the narrowly but the planks are not nearly as wide!)

I didn't do any tests unfortunately but subjectively my cognition was greatly enhanced. (Spent a few years catching up with the insights I had that evening.)

Might be unrelated, but the part about "looped transformers" made me wonder if there's a similar "stepwise" increment going on here.

Edit: Okay, 6.8-18% is slightly less dramatic than what I was referring to.

fc417fc802 14 hours ago|
> I didn't do any tests unfortunately

That's almost always the problem of course. Wasn't there a quote about the "breakthrough" of "shoes go on feet"?

> manipulate them iteratively. (Ordinarily they're more like "glimpses" or "flashes" that fade rapidly.

This is intriguing. I would describe my normal thought process as iteratively working on a semi-persistent problem held in my mind. Is it different for other people?

frunkp 14 hours ago|
When I saw "hidden reasoning", it reminded me of diffusion models: generating a block spans many steps (with remasking), which hide the reasoning that led to the block.

I had not heard of looped transformers, but the engineering behind the number of loops per token / halting feels like trying to apply a diffusion process to a transformer while keeping the auto-regressive feature.

More comments...