Experiments


Here are my ongoing experiments in building tiny, intelligent language models. Each experiment is documented with code, datasets, and results shared openly.

All experiments are reproducible - you can find the code, data, and model weights in the linked repositories.

Sep. 17, 2025

Train a tiny model on Grade School Math

Starmind Pico Dumb model trained on simple math = smarter model?

Aug. 5, 2025

Training a language model to run on RP2040

Starmind Pico Is it possible to train a Language Model to run on the RP2040? Yes. Dumb but fast