Comments URL: https://news.ycombinator.com/item?id=47726788

Points: 4

# Comments: 1

The results are more spectacular...

The model pointed way better in gsm8k, but lost a bit on the other categories.

On the prior art: you're right that layer duplication has been explored before. What I think is new here is the systematic sweep toolkit + validation on standard benchmarks (lm-eval BBH, GSM8K, MBPP) showing exactly which 3 layers matter for which model. The Devstral logical deduction result (0.22→0.76) was a surprise to me.

If there are ComfyUI nodes that do this for image models, I'd love links, the "cognitive modes" finding (different duplication patterns that leads to different capability profiles from the same weights) might be even more interesting for diffusion models.

  # Run lm-evaluation-harness
  lm_eval --model local-chat-completions \
      --model_args model=test,base_url=http://localhost:8089/v1/chat/completions,num_concurrent=1,max_retries=3,tokenized_requests=False \
      --tasks gsm8k_cot,ifeval,mbpp,bbh_cot_fewshot_logical_deduction_five_objects,mbpp \
      --apply_chat_template --limit 50 \
      --output_path ./eval_results

I chat with the model to see if the thing was still working and seemed coherent to me, I didn't notice anything off.

I need to automate testing like that, where you pick the local maxima and then iterate over that picking layers to see if it's actually better, and then leave the thing running overnight

if this is validated enough it can eventually lead to ship some kind of "mix" architecture with layers executed to fit some "vibe?"

Devstral was the first one I tried and optimize for math/eq, but that din't result in any better model, then I added the reason part, and that resulted in "better" model

I used the devstral with the vibe.cli and it look sharp to me, thing didn't fail, I also used the chat to "vibe" check it and look ok to me.

The other thing is that I pick a particular circuit and that was "good" but I don't know if it was a local maxima, I think I ran just like 10 sets of the "fast test harness" and pick the config that gave the most score... once I have that I use that model and run it against the llm_eval limited to only 50 tests... again for sake of speed, I didn't want to wait a week to discover the config was bad

I'm using the following configuration --tasks gsm8k_cot,ifeval,mbpp,bbh_cot_fewshot_logical_deduction_five_objects,mbpp I did also try humaneval but something in the harness is missing and failed...

notice that I'm running 50 tests for each task, mostly because of time limitation as it takes like two hours to validate the run for the base model and the modified one.

I'll also try to publish the results of the small tests harness when I'm testing the multiple layers configurations, for reference this is phi-4-Q6_K.gguf, still running, I'm now giving more importance to the Reason factor, the reason factor comes from running a small subset of all the problems in the task config above

Initially I tried the approach of the highest math/eq but in resulted in models that were less capable overall with the exception of math, and math like in the original research is basically how good was the model at giving you the answer of a really though question, say the cubic root of some really large number... but that didn't translate to the model being better at other tasks...

  Config  | Lyr | Math   | EQ    | Reas   | Math Δ  | EQ Δ  | Reas Δ  | Comb Δ
  --------|-----|--------|-------|--------|---------|-------|---------|-------
  BASE    |   0 | 0.7405 | 94.49 | 94.12% |     --- |   --- |     --- |    ---
  (6,9)   |   3 | 0.7806 | 95.70 | 94.12% | +0.0401 | +1.21 |  +0.00% |  +1.21
  (9,12)  |   3 | 0.7247 | 95.04 | 94.12% | -0.0158 | +0.55 |  +0.00% |  +0.55
  (12,15) |   3 | 0.7258 | 94.14 | 88.24% | -0.0147 | -0.35 |  -5.88% |  -6.23
  (15,18) |   3 | 0.7493 | 95.74 | 88.24% | +0.0088 | +1.25 |  -5.88% |  -4.63
  (18,21) |   3 | 0.7204 | 93.40 | 94.12% | -0.0201 | -1.09 |  +0.00% |  -1.09
  (21,24) |   3 | 0.7107 | 92.97 | 88.24% | -0.0298 | -1.52 |  -5.88% |  -7.41
  (24,27) |   3 | 0.6487 | 95.27 | 88.24% | -0.0918 | +0.78 |  -5.88% |  -5.10
  (27,30) |   3 | 0.7180 | 94.65 | 88.24% | -0.0225 | +0.16 |  -5.88% |  -5.73
  (30,33) |   3 | 0.7139 | 94.02 | 94.12% | -0.0266 | -0.47 |  +0.00% |  -0.47
  (33,36) |   3 | 0.7104 | 94.53 | 94.12% | -0.0301 | +0.04 |  +0.00% |  +0.04
  (36,39) |   3 | 0.7017 | 94.69 | 94.12% | -0.0388 | +0.20 |  +0.00% |  +0.20
  (6,10)  |   4 | 0.8125 | 96.37 | 88.24% | +0.0720 | +1.88 |  -5.88% |  -4.01
  (9,13)  |   4 | 0.7598 | 95.08 | 94.12% | +0.0193 | +0.59 |  +0.00% |  +0.59
  (12,16) |   4 | 0.7482 | 93.71 | 88.24% | +0.0076 | -0.78 |  -5.88% |  -6.66
  (15,19) |   4 | 0.7617 | 95.16 | 82.35% | +0.0212 | +0.66 | -11.76% | -11.10
  (18,22) |   4 | 0.6902 | 92.27 | 88.24% | -0.0504 | -2.23 |  -5.88% |  -8.11
  (21,25) |   4 | 0.7288 | 94.10 | 88.24% | -0.0117 | -0.39 |  -5.88% |  -6.27
  (24,28) |   4 | 0.6823 | 94.57 | 88.24% | -0.0583 | +0.08 |  -5.88% |  -5.80
  (27,31) |   4 | 0.7224 | 94.41 | 82.35% | -0.0181 | -0.08 | -11.76% | -11.84
  (30,34) |   4 | 0.7070 | 94.73 | 94.12% | -0.0335 | +0.23 |  +0.00% |  +0.23
  (33,37) |   4 | 0.7009 | 94.38 |100.00% | -0.0396 | -0.12 |  +5.88% |  +5.77
  (36,40) |   4 | 0.7057 | 94.84 | 88.24% | -0.0348 | +0.35 |  -5.88% |  -5.53
  (6,11)  |   5 | 0.8168 | 95.62 |100.00% | +0.0762 | +1.13 |  +5.88% |  +7.02
  (9,14)  |   5 | 0.7245 | 95.23 | 88.24% | -0.0160 | +0.74 |  -5.88% |  -5.14
  (12,17) |   5 | 0.7825 | 94.88 | 88.24% | +0.0420 | +0.39 |  -5.88% |  -5.49
  (15,20) |   5 | 0.7832 | 95.86 | 88.24% | +0.0427 | +1.37 |  -5.88% |  -4.52
  (18,23) |   5 | 0.7208 | 92.42 | 88.24% | -0.0197 | -2.07 |  -5.88% |  -7.95
  (21,26) |   5 | 0.7055 | 92.89 | 88.24% | -0.0350 | -1.60 |  -5.88% |  -7.48
  (24,29) |   5 | 0.5825 | 95.04 | 94.12% | -0.1580 | +0.55 |  +0.00% |  +0.55
  (27,32) |   5 | 0.7088 | 94.18 | 88.24% | -0.0317 | -0.31 |  -5.88% |  -6.19
  (30,35) |   5 | 0.6787 | 94.69 | 88.24% | -0.0618 | +0.20 |  -5.88% |  -5.69
  (33,38) |   5 | 0.6650 | 94.96 | 88.24% | -0.0755 | +0.47 |  -5.88% |  -5.41
  (6,12)  |   6 | 0.7692 | 95.39 | 94.12% | +0.0287 | +0.90 |  +0.00% |  +0.90
  (9,15)  |   6 | 0.7405 | 94.65 | 94.12% | -0.0000 | +0.16 |  +0.00% |  +0.16
  (12,18) |   6 | 0.7582 | 94.57 | 88.24% | +0.0177 | +0.08 |  -5.88% |  -5.80
  (15,21) |   6 | 0.7828 | 93.52 | 88.24% | +0.0423 | -0.98 |  -5.88% |  -6.86
  (18,24) |   6 | 0.7308 | 92.93 | 94.12% | -0.0097 | -1.56 |  +0.00% |  -1.56
  (21,27) |   6 | 0.6791 | 92.54 | 82.35% | -0.0615 | -1.95 | -11.76% | -13.72

Transformers appear to have discrete "reasoning circuits" — contiguous blocks of 3-4 layers that act as indivisible cognitive units. Duplicate the right block and the model runs its reasoning pipeline twice. No weights change. No training. The model just thinks longer.

The results on standard benchmarks (lm-evaluation-harness, n=50):

Devstral-24B, layers 12-14 duplicated once: - BBH Logical Deduction: 0.22 → 0.76 - GSM8K (strict): 0.48 → 0.64 - MBPP (code gen): 0.72 → 0.78 - Nothing degraded

Qwen2.5-Coder-32B, layers 7-9 duplicated once: - Reasoning probe: 76% → 94%

The weird part: different duplication patterns create different cognitive "modes" from the same weights. Double-pass boosts math. Triple-pass boosts emotional reasoning. Interleaved doubling (13,13,14,14,15,15,16) creates a pure math specialist. Same model, same VRAM, different routing.

The circuit boundaries are sharp — shift by one layer and the effect disappears or inverts. Smaller models (24B) have tighter circuits (3 layers) than larger ones (Ng found 7 layers in 72B).

Tools to find circuits in any GGUF model and apply arbitrary layer routing are in the repo. The whole thing — sweep, discovery, validation — took one evening.

Happy to answer questions.

Comments URL: https://news.ycombinator.com/item?id=47431671

Points: 265

# Comments: 80

[0] https://www.steamdeck.com/en/tech [1] https://www.steamdeck.com/en/tech/deck

  CryptoWorker start b3d84ba5-2f9c-419b-b555-f8d8c4024d5b:1:68170
  CryptoWorker start b3d84ba5-2f9c-419b-b555-f8d8c4024d5b:1:68170
  CryptoWorker start cbab8426-c9ae-44fd-9b77-e6858fc7cfae:1:68170
  CryptoWorker start cbab8426-c9ae-44fd-9b77-e6858fc7cfae:1:68170
  CryptoWorker start e8360a4a-aae4-493c-afa9-ff19753fae18:1:68170
  CryptoWorker start e8360a4a-aae4-493c-afa9-ff19753fae18:1:68170

using Firefox 116 64-bit on ubuntu 23.04

Comments URL: https://news.ycombinator.com/item?id=37596827

Points: 3

# Comments: 0

https://www.amazon.com/Sennheiser-31-II-Binaural-Headset-Mic...

and in case your laptop/desk doesn't have mic and headphone jack you can use

https://www.amazon.com/Sabrent-External-Adapter-Windows-AU-M...

that works with Linux, not sure about windows/mac.

for the cellphone plantronics voyager legend, which is expensive but works every time very well.

As we get older more and more energy would be used on preservation (read it as fix damage and less efficient process as result of age), therefore shrinking/eliminating everything not being used it's necessary.

The second is an entry on how our bodies are machines oriented to try to avoid wasting energy... or better said preserving it... for that part my canonical reference would be the Algernon argument:

http://www.gwern.net/Drug%20heuristics

Edit: TL;DR: From the PDF conslussions:

it is found that working hours up to 25–30 hours per week have a positive impact on cognition for males depending on the measure and up to 22–27 hours for females. After that, working hours have a negative impact on cognitive functioning.

OTOH I see this as the optimization of hardware to fix the cap that software, paradigms and devs mean.

Another way of reading this is that making more use of a given tech should pay the initial cost faster thus making possible more research and improved processes.

Sadly another way of reading this is that Intel doesn't have any tech that would capitalize in better chips anymore.

The solution approach I think goes as back as when using the express card port on elitebooks

https://www.youtube.com/watch?v=IDiizICogMQ

It would be incredible on the other hand to get this solutions as official products from vendors like nvidia or asus without the DIY (not because I'm against it but to push making use of this solutions officially will improve the state of the solutions).

Intel first eliminated the whole aftermarket entry level gpu industry and will probably eliminate the middle tier also.

As Pluma states

"If you're going with a dedicated graphics card in your laptop, battery life is already out of the window, so you might as well get as much processing power as the thing can handle"

Anyone with a youtube link?