Neverduft

Friendly reminder: You don't need to be better than your opponent to trash talk them.

We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…

I've always believed that intelligence should come from simple, general building blocks that learn their own abstractions directly from raw data; in other words, the bitter lesson. I think we're finally seeing a real step towards AI that can learn from all data. Language modeling on raw bytes is a microcosm of this ideal, and to me it's a central problem in AI: if we can't build LLMs that learn end-to-end from bytes, how will we handle the rest of the world's data that we understand even less? Last year we introduced hierarchical networks (H-Nets) as a first step. Our new version finds up to 3-5x efficiency gains in controlled scaling law experiments, and at larger scale matches an open-source model trained on 11T tokens with ~10x less data and compute. The posts below walk through what we found and why we think it works.

Done.

average lucid dream in 2010:

This was significantly harder than I thought it would be.

yeah dude we need flies to hasten the AGI killing everyone. good take dude. RSI just around the corner this is the shit we need. good one. creep

BREAKING 🚨: NASA's Chandra telescope just discovered a new class of cosmic objects Astronomers found 84 of them in 6 galaxies and have no idea what they are

AMEN (OUT NOW)

Patrick is hiding underground from the rain of destruction. I make random scenes like this to fix bugs. This one helped to fix 4.