Nandi Schoots
  • Home
  • Selected work
  • Thesis
  • Contact
  • Thesis
  • Publications
    • Any Deep ReLU Network is Shallow
    • Finding Sparse Initialisations using Neuroevolutionary Ticket Search (NeTS)
    • Low-Entropy Latent Variables Harm Out-of-Distribution Performance
    • Safety Properties of Inductive Logic Programming
    • Dissecting Language Models: Machine Unlearning via Selective Pruning
    • Extending Activation Steering to Broad Skills and Multiple Behaviours
    • Hidden in Plain Text: Emergence & Mitigation of Steganographic Collusion in LLMs
    • Improving Activation Steering in Language Models with Mean-Centring
    • Relating Piecewise Linear Kolmogorov Arnold Networks to ReLU Networks
    • The Propensity for Density in Feed-forward Models
  • Recent & Upcoming Talks
  • Teaching
    • Learn JavaScript
    • Learn Python

Dissecting Language Models: Machine Unlearning via Selective Pruning

Nicholas Pochinkov
,
Nandi Schoots
· 0 min read
PDF
Last updated on Jul 4, 2025
← Safety Properties of Inductive Logic Programming Feb 24, 2021
Extending Activation Steering to Broad Skills and Multiple Behaviours Jan 1, 0001 →

© 2025 Nandi Schoots. This work is licensed under CC BY NC ND 4.0

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.