Jae Hutchinson

sirmyrrh

·

AI & ML interests

None yet

Recent Activity

reacted to AesSedai's post with 👍 1 day ago

Hi all, I'm posting this as sort of an informal notice + poll. I'm down to about 700GB free of HF space and there's MiniMax-M3 on the horizon, plus a couple other models I'd like to quant like the Nex-N2 Pro finetune. I've already super-squished all of my quant repositories to free up any LFS space that might have been lingering there, but I'm back near the cap again now. To free up some space, I'm planning to remove these three older GLM quants: - GLM-4.5: 1.23TB - GLM-4.6: 728GB - GLM-4.7: 787GB I'm open to other suggestions as well, and I'll wait a few days before removing anything in case someone wants to download a version before I get rid of them. Thanks! Edit 06/14/2026: I've created an organization https://huggingface.co/QuantPasture and moved the older GLM quants there. My GLM-4.6-Derestricted has migrated to ReadyArt: https://huggingface.co/ReadyArt/GLM-4.6-Derestricted-GGUF

reacted to grimjim's post with ➕ 1 day ago

I think it's clear in retrospect that "frankenmerges", which repeated blocks of layers, amounted to a crude approximation of looped transformers architecture, hence them able to work at all instead of just breaking. They lucked out due to much of the signal passing through residual streams being preserved and only modulated along the way. That said, not all models are suited for this. Models which feature ever-increasing magnitudes as inference progressess through layers risk exploding precision limits.

liked a model 3 months ago

Blazed-Forge/Gemma-4-Sphinsikus-Chronist-31B

View all activity

Organizations

models 2

sirmyrrh/Kyllima-34B-v1

Text Generation • 34B • Updated Oct 15, 2024 • 41

sirmyrrh/Kyllima-34B-v1-GGUF

34B • Updated Oct 3, 2024 • 68

datasets 0

None public yet