Entries, Model Internals | Paula Livingstone

Skip to content

Paula Livingstone writing · projects · tools

Home Writing Tools Projects Thoughts About Giving Back Contact

Projects

Model Internals

A hands-on, evidence-first investigation into what actually happens inside a running language model, opening one up on local hardware and testing directly why it does what it does.

PythonPyTorchHugging Face TransformersQwen3-4BApple M4 ProMetal

Overview How AI Agents Decide and Act Teaching a Model to Use Tools Forcing Valid Output Getting Reliable Structured Answers Giving a Model Outside Knowledge How Conversation Memory Works Fine-Tuning Without Retraining Turning Meaning Into Numbers Reading a Model's Confidence Controlling Randomness in Output

Overview How AI Agents Decide and Act Teaching a Model to Use Tools Forcing Valid Output Getting Reliable Structured Answers Giving a Model Outside Knowledge How Conversation Memory Works Fine-Tuning Without Retraining Turning Meaning Into Numbers Reading a Model's Confidence Controlling Randomness in Output

Turning Meaning Into Numbers

0 entries

No entries in this section yet. They’re on the way.

Paula Livingstone

Writing, projects, tools, and selected work on OT security, AI governance, and systems engineering.

Explore

Writing Tools Projects Thoughts

Connect

About Contact Pay

© 2026 Paula Livingstone

Back to top ↑