Ember is a memory-efficient optimizer for token embeddings that exploits distinct gradient geometry and models token trajectories as 1D rays.
Muon: An optimizer for LLMs.Keller Jordan’s Blog, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Token Geometry
Ember is a memory-efficient optimizer for token embeddings that exploits distinct gradient geometry and models token trajectories as 1D rays.