Rasmus Larsen
ML research engineer at
Danish Foundation Models
in Copenhagen
Writing
Hurry Up And Wait: The Life of an RL Sandbox (Part 1)
Soon
Quantifying LLM Base Model Quality through Compression
Apr 2025
Projects
maester
↗
Language model training framework in modern PyTorch
verifiers-ucloud
↗
UCloud sandbox runtime for RL environments built with verifiers
Elsewhere
GitHub
↗
rlrs
X
↗
@synquid
Hugging Face
↗
synquid