← Înapoi la Urmăritori

Cel mai recent articol

Ultima verificare

Urmăritori 0 urmăritori
Articole: 7
  1. Most of the time at inference, the Fisher pullback through the unembedding tells you exactly how far you can perturb the residual stream before the predictive distribution moves — and it's literally the second-order Taylor expansion of KL.

  2. A v1.0 rewrite of my old curvature analysis library: new operators, new algorithms, and a lot of numerical validation I should have written the first time.

  3. Understanding how attention can express logical propositions about correlations between tokens and other fun tips and tricks

  4. Understanding how attention can express logical propositions about correlations between tokens and other fun tips and tricks

  5. Fine-tuned noise

    Studying the covariance structure of mini-batch noise in stochastic gradient descent

  6. MAML Hunting

  7. An application of dynamical systems theory to optimization