Blogs

  • Loss Landscapes and the Hessian Matrix
    February 2025
    A look at how we can understand loss landscapes and the associated Hessian. Read on.
  • On LLM Internals and Unlearning
    February 2025
    A (very) brief overview on interpretability and manifold learning. Also, my thoughts on how machine unlearning may reveal newer strategies to expose LLM internals. Read on.