The complete logs of every experiment in the blog post Sparks of Weight Space Programming: LLMs can understand and write neural network weights. For each experiment you get: everything we sent the ...

In the strongest example of "weight space programming", as the technique is called, Astra’s intervention raised Qwen3.5-0.8B’s MATH-500 score from 7.9% to 48.1% while touching approximately 0.016% of ...

Astra identifies game policies from raw weight matrices and fixes reasoning loops in Qwen3.5-0.8B by editing 0.016% of parameters, lifting MATH-500 scores from 7.9% to 48.1% without training loops.

Article by Shenchao TechFlowNo training required, no data needed, no gradient descent.Researchers at MIT fed the weights of ...

A pair of tiny neural networks can generate and play five simple games on a Raspberry Pi Pico 2 W without conventional game ...

MIT CSAIL researchers say GPT-6 Astra edited 0.016% of Qwen3.5-0.8B weights to fix reasoning loops, lifting MATH-500 from 7.9% to 48.1% without training.

What happens when you ask ChatGPT to recreate Snake, Tetris, Pong, Pac-Man and Space Invaders? I tested its new game-making ...

A new MIT CSAIL blog post is making the rounds in AI circles this week, and it describes something that sounds almost ...

TinyWorldFormer uses compact neural networks and a looped transformer to generate playable games and train an AI agent ...

Sign up for Ordinary Extraordinary, Ian Bogost’s guide to making everyday life vivid again. My family and I enjoy the occasional meal of delicious slow-cooked meat ...

A language model that does not generate text. It reads a state and a set of typed questions and returns, from one forward pass, a probability distribution for every question. Recorded episodes; every ...