Inversion Lab for AI Safety

AI & ML interests

Interpretability of Language Models and Multi-Agent Safety

Recent Activity

lgalke  authored a paper 2 days ago

Selecting The Most Informative Tokens in Natural Language Autoencoders

lgalke  authored a paper 2 days ago

Write Once, Run Everywhere: The Axon DSL for Shape-Safe and Framework-Agnostic LLM Architectures

lgalke  authored a paper 2 days ago

DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible Post-Training Data

View all activity

Team members 1