[com2kid]

Anyone remember that blog post from a few months back where someone was able to improve a model's math ability by just duplicating layers that were activated while solving math problems? Just literally copy/pasting them and linking them together so the model ran through the same layers again?

I get the feeling a lot more research is going to come out in the area of exploring exactly what portions of a model's weights do what.

[logancbrown]

Source for those interested

https://dnhkng.github.io/posts/rys/

[jimbokun]

Great, clear write up! Made it very easy to understand.

[marshray]

If dirt-simple type operations like copy-paste yield useful improvements with even a small probability that would seem to open things up for adaptive reconfiguration and whole other classes of optimizations like genetic algorithms.

[artful314w]

That right. Here is some interesting research on that https://sakana.ai/evolutionary-model-merge/

[wongarsu]

Found it: https://news.ycombinator.com/item?id=47500709

Part 3 might be the best introduction: https://dnhkng.github.io/posts/sapir-whorf/

tl;dr: Based on experiments with similar prompts translated to different languages LLM layers group into three phases: the first decodes from the source language into an abstract space, the middle does something, then there's a last part where the abstract result gets transformed back to the target language. And you can repeat the middle to get a stronger model. Which neatly fits Anthropic's findings here that something similar to CoT is happening in those middle layers

Three months ago. I wonder if Anthropic's J-Space research was actually inspired by those blog posts

[mike_hearn]

Nah, it's a cool blog post especially as it was real AI research done at home (albeit with a ridiculously expensive PC), but Anthropic and other labs have been investigating this kind of thing for years.

Even the original transformer architecture makes this clear. It had an explicit "encoder" phase and then a "decoder" phase. Modern LLMs collapse the two together, or are sometimes described rather confusingly as being decoder only. But what they're doing is more or less the same.

[dnhkng]

Author here:

Yeah, the encoder and decoder stuff is explicit, but the internal structure in generated during training. I don't think the big labs were doing this back when I did the research; no one was back in '24.

I just didn't get round to publishing for years, because I have a day job.

By the way, it still works! I tested it earlier this year on Qen3.6 and you still see improvements, so either a) no one actually paid attention, or b) it has more room to scale.

[mike_hearn]

I think you're right that the idea of looping layers is unique to you, congrats and thanks for writing those great blog posts (I read them for the first time a few days ago!). But the idea that the thinking is happening in an abstract space via neural circuits in the middle layers I feel was one that I was reading about in 2024 at least, as Anthropic have been doing this kind of research for a long time. Maybe I'm misremembering though!

My impression from reading the literature is that there are a gazillion interesting ideas and findings published that nobody is picking up in production models. The big labs are researcher constrained, there just aren't enough hours in the day to keep up with the literature and integrate all the interesting ideas found there. So it's not surprising that your trick still works. It'd be even less surprising to discover nobody at these labs has read your blogs, or they have but never found time to experiment with them. Or, they tried, but there is no set of loops that improves some metrics without harming others - I would expect neural circuits to be misaligned across the middle layers so looping layers for one task would put a fault line in circuits for other tasks.

Then they have to trade off the extra GPU capacity needed to do the extra layers, and so on.

[tedd4u]

The Ouro looping results are interesting [1] and they are focused more on the improved reasoning from looping middle layers rather than the parameter efficiency aspect. They train 1.4 and 2.6B parameter models with 7T tokens. The training includes learning how many times to loop on any given token (there’s an early exit module). My guess as to why (as far as we know) looping is not in frontier models yet is that, at frontier training run scale, it’s probably going to require a lot of trial and error and at-scale research. While currently they already probably have a list of dozens or hundreds of of promising ideas that don’t complicate things as much. In the other hand, Ouro’s looping technique shows ability to compete well with models with 3x parameters which seems attention-getting to me. If there’s another 3x to be had down that path. It’s order of magnitude opportunity. Btw there is a great related work section in the paper.

[1] https://ouro-llm.github.io/

[DoctorOetker]

I wonder if the dnhkng results could be correlated to reasoning in first order logic / set theory notation?

There will be multiple notations (MetaMath, Lean, and essentially Frege's notation everyone learns in high school), and we could try to identify how the neural networks represent them as vectors (or vector combinations). The moment formal logic can be connected to the reasoning representations, regularization can be reduced to eliminating internal inconsistencies.

[gwerbin]

That's a cool result because that's also kind of what's happening inside the transformer unit: project -> QKV fuzzy lookup -> unproject. And in a different direction, it's analogous in some sense to what's happening in stacked convolutional layers, where the layers at different levels learn to recognize features of increasing detail.

[wolttam]

Yeah! I still think about that sometimes. Mind-blowing that worked at all, let alone improved performance.

[echelon]

> I get the feeling a lot more research is going to come out in the area of exploring exactly what portions of a model's weights do what.

Too bad the frontier models are closed weights.

Maybe the research community and whole rest of the world will build on open and all the advances will happen in open ecosystems instead.

[ayewo]

A Google DeepMind researcher (Neel Nanda) was able to replicate their claims on an open weight model (Qwen 3.6 27B):

> We have replicated the core claims on Qwen 3.6 27B, and also share preliminary evidence of extending this work by finding abstract "interpretative meta-tokens", like Chinese characters for "what does this mean" that seem to activate and play a causal role on processing ambiguous sentences

See p33 of [1]

Anthropic also released companion code to go with their paper in [2] which also used Qwen. They state that their code should be broadly adaptable to other open weight models with HuggingFace decoders.

[1]: https://www-cdn.anthropic.com/files/4zrzovbb/website/cc4be24...

[2]: https://github.com/anthropics/jacobian-lens

[dnhkng]

Too bad they also don't give anything back to individual researchers. Oh well, wasn't expecting much.

[tuvix]

I always thought that area of research had the coolest name, too: “mechanistic interpretability”

[dr_dshiv]

“Machine psychology” sticks with me. So Asimov.

[throwitaway222]

Worried person cure: Stop overthinking it!

LLM -> AGI fix: START OVERTHINKING!

[DoctorOetker]

it makes you wonder if it may be more efficient to spend all the weights on one layer, and have a repeating stack of the same layer, one would presume this axis has already been explored with metaparameter sweeps?

[smallerize]

That's called a recurrent neural network (RNN).

[DoctorOetker]

distinction is that GPT allows a lot of parallel computation compared to RNN, but I see how your remark indicates a convergence towards RNNs indeed.