Steer the model

Compose a feature-injection formula and watch the model’s answer change — no weight edits, just vectors added to the residual stream while it generates.

Waking the model…

The demo runs on a GPU that scales to zero when idle, so the first visit spins it up and loads the 1.4B model and its six sparse autoencoders into memory. This usually takes about 30–60 seconds; it stays responsive afterward.