René's URL Explorer Experiment


Title: Understanding Muon

direct link

Domain: www.lakernewhouse.com

Links:

Chapter 1: Into the Matrixhttps://www.lakernewhouse.com/writing/muon-1
Chapter 2: Source Codehttps://www.lakernewhouse.com/writing/muon-2
Chapter 3: Weight Regulationhttps://www.lakernewhouse.com/writing/muon-3
Learn with LLM https://www.lakernewhouse.com/muon/llms.txt
(what's this?) https://www.youtube.com/watch?v=LCEmiRjPEtQ&t=2059s
Laker Newhousehttps://www.lakernewhouse.com/
Researchhttps://www.lakernewhouse.com/research
Writinghttps://www.lakernewhouse.com/writing
~ How to MIT ~https://www.lakernewhouse.com/how-to-mit
Chapter 1: Into the Matrix (speedy Muon explainer based on common confusion points)https://www.lakernewhouse.com/writing/muon-1
Chapter 2: Source Code (annotated PyTorch implementation + usage advice)https://www.lakernewhouse.com/writing/muon-2
Chapter 3: Weight Regulation (MuonClip, our work, and future directions)https://www.lakernewhouse.com/writing/muon-3
Jianlin Su's recent blog on QK-cliphttps://kexue.fm/archives/11126
Check out the paper (joint with fantastic collaborators) → https://arxiv.org/abs/2507.13338
"Deriving Muon" by Jeremy Bernstein (why Muon is the way it is)https://jeremybernste.in/writing/deriving-muon
"Why we chose Muon" by Jianlin Su (more philosophical focus)https://kexue.fm/archives/10739
"Muon" by Keller Jordan et al. (more speedrun focus)https://kellerjordan.github.io/posts/muon/
many brancheshttps://docs.modula.systems/intro/reading-list/
Shampoohttps://arxiv.org/abs/1802.09568
spectral descenthttps://proceedings.mlr.press/v38/carlson15.pdf
duality descenthttps://arxiv.org/pdf/1708.00523
PhD thesishttps://arxiv.org/abs/2210.10101
spectral norm workhttps://arxiv.org/abs/2310.17813
modular norm workhttps://arxiv.org/abs/2405.14813
optimizer anthologyhttps://arxiv.org/abs/2409.20325
CIFAR-10 speedrunhttps://github.com/KellerJordan/cifar10-airbench
NanoGPT speedrunhttps://github.com/KellerJordan/modded-nanogpt
large batcheshttps://arxiv.org/abs/2505.02222
up to 1T parametershttps://moonshotai.github.io/Kimi-K2/
Dionhttps://arxiv.org/abs/2504.05295
Polar Expresshttps://arxiv.org/abs/2505.16932
Check out my thesis → https://www.lakernewhouse.com/thesis.pdf
(Anthology, Story 1)https://arxiv.org/pdf/2409.20325
Anthology, Story 2https://arxiv.org/pdf/2409.20325
Chapter 2: Source Code → https://www.lakernewhouse.com/writing/muon-2

Viewport: width=device-width, initial-scale=1.0


URLs of crawlers that visited me.