Research Documents

• Konstantinos Abstract 02 05 26
• SAE Metrics Overview
• SynthSAEBench concepts explained
• connection SynthSAEBench Universality
• evaluating SAEs
• key Papers Summary
• manifold SAE research plan
• manifold SAE research proposal
• manifold implementation explanation
• paper semantic synthsaebench
• semantic constraints synthsaebench
• semantic hierarchies for safety
• semantic vs statistical hierarchies
• session_notes
• universality SAEbench notes
• universality SAEbench research plan
• zotero mcp manifold safety