Dark | Light
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![cloneofsimo Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1529156127006392320.png) Simo Ryu [@cloneofsimo](/creator/twitter/cloneofsimo) on x 14.5K followers
Created: 2025-07-12 22:35:13 UTC

MuonClip... so many tricks to make maximum logits bounded during training. Gets me wondering why dont people try LASER (and maybe, z-loss ?)


XXXXXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1944163666200604934/c:line.svg)

**Related Topics**
[gradient](/topic/gradient)

[Post Link](https://x.com/cloneofsimo/status/1944163666200604934)

[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

cloneofsimo Avatar Simo Ryu @cloneofsimo on x 14.5K followers Created: 2025-07-12 22:35:13 UTC

MuonClip... so many tricks to make maximum logits bounded during training. Gets me wondering why dont people try LASER (and maybe, z-loss ?)

XXXXXX engagements

Engagements Line Chart

Related Topics gradient

Post Link

post/tweet::1944163666200604934
/post/tweet::1944163666200604934