AIBussinBuild useful systems with modern AI
  • Home
  • Books
  • Solutions
  • Toolkit
  • Prompts
  • Articles
  • About

Gradients

  • What Are We Actually Doing?
  • Autograd: What Did PyTorch Record, and Where Does the Gradient Stop?
  • Training: Which Link in the Learning Chain Is Broken?
  • PACS — Building an Optimizer From Gradient Statistics
  • PyTorch Model Not Learning? A Systematic Debugging Guide
  • PyTorch Autograd Debugging: requires_grad, detach, backward() and NaN Gradients
AIBussin

Build useful systems with modern AI.

Books, solutions, experiments and working tools for practical AI.

© 2026 Ernan Hughes
  • Books
  • Solutions
  • Toolkit
  • Prompts
  • Articles
  • About
  • Programmer.ie
  • ZeroModel.org
  • GitHub