CourseTransformers: From Attention to Trained, Served and Interpreted Models · Module 4: Positional Information · part 31 of 111
Part 31 · Module 4: Positional Information

Lesson 4: Position on a grid, a graph, or nowhere

20 min read·13 Oct 2026
Module 4 · Lesson 4 · About 45 minutes · Needs: Module 4, Lessons 1 and 2 (sinusoidal_table, rope, permutation equivariance)

The situation

Isla Duarte, the robotics engineer, has seen the position switch. Her robots read 32 by 32 camera test cards, and an arm must know where an object is before it reaches for it. "Your positions are a line," she says. "My cards are a grid. Does your lab even know the difference?"

This lesson gives tinyformer/positions.py grid positions and a test for whether a model uses them. Module 13 builds the full vision transformer; here you build only enough of one to measure position.

The rest of this course is yours to keep

This course is bought on its own, once, and stays readable afterwards, including the parts added to it later.