Jonathan Dunlap

Life is a temporary celebration of something larger and eternal. We walk in the shadows of countless generations before us that gave us this shared reality to make our own. Human wealth is not measured in abstract numbers.. but in experience, love, and in new ideas imagined.🍄

That water clarity is an engineering decision, and the math behind it is wilder than the video. Roman aqueducts ran on gravity alone. No pumps, no pressure systems. Engineers carved channels with a gradient so shallow it borders on absurd. The Pont du Gard in southern France drops 2.5 centimeters over 275 meters. That's roughly the thickness of a coin over the length of three football fields. They surveyed that accuracy with plumb lines and wooden leveling instruments. The clarity you're seeing is a direct product of flow velocity. Too steep and the water erodes the channel walls, picks up sediment, turns brown. Too flat and it stagnates. Roman engineers targeted a slope of about 20 centimeters per kilometer, which kept the water moving fast enough to stay fresh but slow enough to stay clear. Before the water reached the city, it passed through multi-chamber settling tanks where velocity dropped near zero. Suspended particles sank. Clean water flowed out the top into the next chamber. Repeat three or four times. Pliny specified the minimum slope in writing. Vitruvius published the exact mortar ratio for hydraulic cement: one part lime to two parts volcanic ash for underwater work. The pozzolana from Pozzuoli reacted with water to form a calcium-aluminum-silicate compound that actually gets stronger the longer it sits submerged. Modern concrete degrades in water. Roman concrete bonds with it. Scale the whole system and it gets harder to process. Eleven aqueducts fed Rome at its peak. Combined output: roughly 1 million cubic meters of water per day. That works out to about 250 gallons per person for a city of one million. Modern New York delivers about 125 gallons per person per day. Ancient Rome had access to double the per capita water supply of the largest city in the United States, running entirely on slope and stone. The Trevi Fountain in Rome is still fed by one of them. Two thousand years, same source, same gravity, same water.

@andrewbowser This character pivot is.. something else. Do more please.

Pastor Phil VS. Charlie (ROUND 1)

Source: businessinsider.com/steve-wozniak-…

Boos: "AI is a rocket with super powers and many people will be able to board it." Cheers: "People have super powers, and AI will be our copilot to improve our capabilities."

@stenichele I just ask as I am fascinating blending 'game of life' with language models, and then watch how different 'language games' play out. I ran lots of neat experiments, but none of them with interest results to post about.

@JonathanRoseD Sorry this is another plot measuring semantic similarity with neighboring cells

Small experiment on "semantic artificial life". I put small LLMs (Qwen 2.5 0.5B) on a 1D lattice. It is an NCA where update rules are LLMs that receive tokens from left/right neighbors & their previous output and produce new output tokens.

@stenichele So the entire 1d field (every cell updated) with a new generation of each new generation (containing self + L/R neighbors)? What is the 'cosine similarity' in the graphs? What's the idea/point of this?

@JonathanRoseD correct, used both as input together with the previous output of the cell

People underestimate the power of shaders! 😎 Most environments look artificial cause the surfaces feel flat and lifeless So, inspired by Forza Horizon, I created this shader to add depth and atmosphere! Completely changing how flat surfaces are perceived Coming soon on Patreon #VRChat #Shader

[1/3] Excited to share Winfree Oscillatory Neural Network (WONN): a synchronization-based neural architecture built on Winfree dynamics. WONN evolves representations through oscillatory synchronization on a toroidal phase space (S^1)^d. Project page: jiawen-dai.github.io/WONN_Project_P…

Lissajous curve table. Best geometry gif ever.

Probabilistic Tiny Recursive Model is a nice example of inference as stochastic fixed-point search (Deep Manifold Part 2: Neural Network Mathematics, arXiv:2512.06563) Deterministic recursion follows one trajectory and can get trapped in a bad basin; weak latent perturbation opens nearby convergence pathways. From the Deep Manifold lens, this is not random noise destroying structure: small admissible perturbation preserves local homology of the manifold cover while shifting the trajectory among fixed-point classes. The Q head then acts like a learned basin-quality functional, selecting the better fixed point. This is why stochastic width can outperform simply adding more deterministic depth.

The HRM-Text paper is now available 🎉 HRM-Text explores a different approach to language model pretraining: hierarchical recurrent computation, task-completion training, and latent-space reasoning. At just 1B parameters, HRM-Text achieves competitive performance with dramatically lower training cost and data requirements. 1B parameters 40B unique tokens ~1 day of pretraining ~$1000 training cost

@GeminiApp @Google @antigravity @GoogleAIStudio This is just the start – Gemini 3.5 Pro is coming next month. Find out more ↓ goo.gle/3PwoinZ #GoogleIO

@IlkanAkten @UnslothAI That's what I've heard too. The MTP isn't that notable for MoE models.

@UnslothAI I see almost no difference between standard and mtp of 3.6 35B a3b on 9800x3d, 64gb ram and 3080 10gb. One is 49t/s, mtp is 53t/s

4-bit Qwen3.6 MTP GGUF managed to search 70+ sites from a single prompt. Try this locally on 20GB RAM via Unsloth Studio. Unsloth now supports auto MTP + speculative decoding & auto-selects the best MTP settings for your device (Mac, CPU, GPU). GitHub: github.com/unslothai/unsl…

there's no catch; SAM3 is open source and really good one of the things it does really well is object tracking, even in crazy complex scenes like basketball probably my favorite computer vision model ever

The longevity dose for sleep is 6.4 - 7.8 hours. > 23 biological aging clocks > multi-omics: 11 proteomic, 5 metabolomic, 7 MRI > 500,000 people Interesting findings: + Brain proteins notice sleep loss before brain anatomy does. When you measure brain aging by plasma proteins, the brain looks biologically youngest at 7.82 hours of sleep in women and 7.70 hours in men. When you measure brain aging by MRI of brain anatomy, it looks youngest at 6.48 hours in women and 6.42 hours in men. + The brain and the metabolic organs share the same U-shape but hit their optimum at different hours. Fat tissue and the pancreas both bottomed near 6 hours. The brain bottomed higher, between 6.4 and 7.8 hours depending on whether you measure by MRI or plasma proteins. Sleep less or more than the organ-specific optimum and aging accelerates. + Short sleepers vs long sleepers DNA. Short sleepers' DNA matched the DNA of people whose bodies are breaking down all over. > back pain 40% > depression 37% > substance use disorders 37% > anxiety 32% > heart failure 31% > lung disease 28% > type 2 diabetes 18%. Looking at genes only, chronically too little sleep makes the body look like it's breaking down everywhere. Long sleepers' DNA matched the DNA of people with brain conditions versus whole body breakdown. > major depression 29% > schizophrenia 28% > ADHD 28%, > migraine 28% > bipolar disorder 21% Short sleep gets you through the body directly: the nervous system get's aggravated, the immune system gets confused and stress hormones flood the bloodstream. Long sleep get's you through the brain, but it's the result and not the cause. By the time someone is sleeping too long, the damage is already happening inside their organs. Summary: Less than 6.4 hours is a stressor. Your body is wearing down because it never gets enough time to recover. The short sleep is what is causing the damage. More than 7.8 hours is a warning sign, signaling that something is already going wrong in your brain or your metabolic organs.

llama.cpp with MTP support makes local models fast enough to use as daily drivers 🚀 Qwen3.6-27B dense generation (on A10G): From 25 tok/s → 45 tok/s (+78%). Two flags on llama-server: --spec-type draft-mtp --spec-draft-n-max 2

@PNASNexus That's because space is a non-Newtonian fluid. The vacuum of space is not truly empty and is teeming with quantum fields. Because quantum vacuum fluctuations exhibit pressure and energy, some theoretical frameworks treat the vacuum mathematically like a fluid or a superfluid.

The gravitational dynamics that shape entire galaxies, creating tidal arms and bridges, can be reproduced at the millimeter scale by water lenses on a soap film—the vast and slow ballets of space are echoed in miniature on an ephemeral soap bubble: ow.ly/sm5h50YZNQR

I strongly believe there are entire companies right now under heavy AI psychosis and its impossible to have rational conversations about it with them. I can't name any specific people because they include personal friends I deeply respect, but I worry about how this plays out. I lived through the great MTBF vs MTTR (mean-time-between-failure vs. mean-time-to-recovery) reckoning of infrastructure during the transition to cloud and cloud automation. All those arguments are rearing their ugly heads again but now its... the whole software development industry (maybe the whole world, really). It's frightening, because the psychosis folks operate under an almost absolute "MTTR is all you need" mentality: "its fine to ship bugs because the agents will fix them so quickly and at a scale humans can't do!" We learned in infrastructure that MTTR is great but you can't yeet resilient systems entirely. The main issue is I don't even know how to bring this up to people I know personally, because bringing this topic up leads to immediately dismissals like "no no, it has full test coverage" or "bug reports are going down" or something, which just don't paint the whole picture. We already learned this lesson once in infrastructure: you can automate yourself into a very resilient catastrophe machine. Systems can appear healthy by local metrics while globally becoming incomprehensible. Bug reports can go down while latent risk explodes. Test coverage can rise while semantic understanding falls. Changes happens so fast that nobody notices the underlying architecture decaying. I worry.