OpenRocket

Austin’s surge of new housing construction drove down rents

A sufficiently detailed spec is code

Afroman found not liable in defamation case brought by Ohio cops who raided home

As an experiment I started asking Claude to explain things to me with a fiction story and it ended up being really good, so I started seeing how far I could take it and what it would take to polish it enough to share publicly.Over the last couple months, I&#x27;ve been building world bibles, writing and visual style guides, and other documents for this project… think the fiction equivalent of all the markdown files we use for agentic development now. After that, this was about two weeks of additional polish work to cut out a lot of fluff and a lot of the LLM-isms. Happy to answer any questions about the process too if that would be interesting to anybody.

Warranty Void If Regenerated

Nvidia greenboost: transparently extend GPU VRAM using system RAM/NVMe

Nvidia NemoClaw

Show HN: I built 48 lightweight SVG backgrounds you can copy/paste

Wander – A tiny, decentralised tool to explore the small web

Cook: A simple CLI for orchestrating Claude Code

Conway's Game of Life, in real life

Astral to Join OpenAI

I replicated David Ng&#x27;s RYS method (<a href="https:&#x2F;&#x2F;dnhkng.github.io&#x2F;posts&#x2F;rys&#x2F;" rel="nofollow">https:&#x2F;&#x2F;dnhkng.github.io&#x2F;posts&#x2F;rys&#x2F;</a>) on consumer AMD GPUs 
(RX 7900 XT + RX 6950 XT) and found something I didn&#x27;t expect.Transformers appear to have discrete &quot;reasoning circuits&quot; — 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 degradedQwen2.5-Coder-32B, layers 7-9 duplicated once:
- Reasoning probe: 76% → 94%The weird part: different duplication patterns create different cognitive &quot;modes&quot; 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.

Show HN: Duplicate 3 layers in a 24B LLM, logical deduction .22→.76. No training

HN

OpenRocket

Austin’s surge of new housing construction drove down rents

A sufficiently detailed spec is code

Afroman found not liable in defamation case brought by Ohio cops who raided home

Warranty Void If Regenerated

Nvidia greenboost: transparently extend GPU VRAM using system RAM/NVMe

Nvidia NemoClaw

Show HN: I built 48 lightweight SVG backgrounds you can copy/paste

Wander – A tiny, decentralised tool to explore the small web

Cook: A simple CLI for orchestrating Claude Code

Conway's Game of Life, in real life

Astral to Join OpenAI

Show HN: Duplicate 3 layers in a 24B LLM, logical deduction .22→.76. No training

The math that explains why bell curves are everywhere

Autoresearch for SAT Solvers

Iran war energy shock sparks global push to reduce fossil fuel dependence

2% of ICML papers desk rejected because the authors used LLM in their reviews

LotusNotes

RX – a new random-access JSON alternative

Afroman Wins Civil Trial over Use of Police Raid Footage in His Music Videos