Interactive labs
Learn by changing the system.
These small experiments pair with deeper field notes. Adjust a variable, form a prediction, and watch the behavior change in real time.
Live concept
loss = 1.0x²
Tune gradient descent
Change the learning rate and iteration count. Watch the optimizer settle, oscillate, or overshoot the minimum.
Convergingloss 0.014
The optimizer is moving efficiently.
Paired with the narrative
Labs can render directly in a post, so readers never lose the conceptual thread.
Authored from the CMS
Add an interactive-concept code block in the editor and preview the exact public result.