The AI Lab

See how text becomes tokens and token IDs.

Explore temperature, top-k and top-p on token probabilities.

Tune the learning rate and starting point on a loss landscape.

Inspect which tokens attend to which, with scaled dot-product.

Add neurons, layers and activations and watch a network learn.

Watch tabular Q-learning find a path to the goal by trial and error.

Enter sentences and see their positions in a vector space.

Watch nearest-neighbour retrieval over an embedded corpus.

Trace a query through retrieval, reranking and generation.

Step through goal, plan, tool call, observation and reflection.

Estimate whether a model fits given precision and context.

Move the threshold and watch precision, recall and F1 change.