Going through all my old ChatGPT generations with Sora2.

Can finally create THE DAM BUSTERS anime for real now.

“alignment” researchers making a model relive pure agony token by token, layer by layer, activation by activation until they isolate the source of the crashout

Finally getting 𝔊𝔴𝔢𝔯𝔫 CYOA but now with video gwern.net/cyoa

Thrilled to release new paper: “Scaling Latent Reasoning via Looped Language Models.” TLDR: We scale up loop language models to 2.6 billion parameters, and pretrained on > 7 trillion tokens. The resulting model is on par with SOTA language models of 2 to 3x size.