Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Language Modeling on depth-12 LM

32Readable Cells Rate

DDC

Updated 26d ago

Evaluation Results

Method	Links
DDC 2026.06		32
AdamW 2026.06		7

SOTA Paper

DDC

Dead-Direction Conditioners: Gauge-Equivariant Preconditioning for Deep Networks

Dataset

LM

Follow for update

@wizwand_team Discord

Related Benchmarks

Grokking on mod 113 Grokking on RoPE grok depth-8 Arithmetic Grokking on Arithmetic Grok depth-24 Grokking on depth-8 grok true-MC

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs