Post

New Paper Alert! Can we precisely erase conceptual knowledge from LLM parameters? Most methods are shallow, coarse, or overreach, adversely affecting related or general knowledge. We introduce🪝𝐏𝐈𝐒𝐂𝐄𝐒 — a general framework for Precise In-parameter Concept EraSure. 🧵 1/

Large language models excel at storing knowledge, but not all of it is safe or useful - e.g. chatbots for kids shouldn’t discuss guns or gambling. How can we selectively remove inappropriate conceptual knowledge while preserving utility? Meet our method🪝𝐏𝐈𝐒𝐂𝐄𝐒!

🪝𝐏𝐈𝐒𝐂𝐄𝐒 works by: 1️⃣ Disentangling model parameters into interpretable features (implemented using SAEs) 2️⃣ Identifying those that encode a target concept 3️⃣ Precisely ablating them and reconstructing the weights No need for fine-tuning, retain sets, or enumerating facts. 2/

We show that 🪝𝐏𝐈𝐒𝐂𝐄𝐒: ✅ Achieves much higher specificity and robustness ✅ Maintains low retained accuracy (as low or lower than other methods!) ✅ Preserves coherence and general capabilities 3/

We also check robustness to relearning: Can the model relearn the erased concept from related but non-overlapping data to the eval questions? 🪝𝐏𝐈𝐒𝐂𝐄𝐒 resists relearning far better than prior methods, while others often fully recover the concept! 5/