“Just by looking at the raw weights, GPT-6 Astra can tell what a network was trained to do.”

rikiyat.github.io · Nov 1 · Researcher shows OpenAI's GPT-6 Astra hand-editing Qwen's weights without training

“With the same sampling settings, this weight correction raises MATH-500 from 7.9% to 48.1% and generalizes out of the box to instruction following, raising IFEval by 3x.”

rikiyat.github.io · Nov 1 · Researcher shows OpenAI's GPT-6 Astra hand-editing Qwen's weights without training