ai safety
-
Geometry of Hidden State
-
Causal Attribution & Patching (Part 2): Pathscopes
-
Causal Attribution & Patching (Part 1): Internal Mechanisms of Language Models
-
Training Data Attribution: Corroborative Methods
-
Training Data Attribution: Contributive Methods
-
Beyond the Leaderboard: New Frameworks for Understanding LLM Behavior
-
Inner Workings of a Transformer: Interpretability Techniques
Notes on the “Primer on teh Inner workings of the Transformer based LMs”
-
Part 4 Reverse Engineering Transformers: Induction Heads
-
Part 3 Reverse Engineering Transformers: Path Expansion
-
Part 2: Reverse Engineering Transformers: Attention Heads & Circuits