LINKED LIST [txt mode] ▸ Mixture-of-Depths: Dynamically alloca...
home explore | log in

Mixture-of-Depths: Dynamically allocating compute in transformers

news.ycombinator.com · 2026-10-06 · 0 upvotes

log in to save, upvote or flag this.


─── In 0 lists ─────────────────────────────────────────

(not in any lists yet)

* Mixture-of-Depths: Dynamically allocating compute in transformers (discussion)