NeuroBait: I fine-tuned a model to spark dopamine for ADHD brain
A Gemma 3 fine-tune that helps ADHD users start tiny, meaningful actions.
NeuroBait is a Hugging Face community project built to help with ADHD task-initiation freeze rather than diagnosis or to-do planning. It fine-tunes google/gemma-3-12b-it with LoRA to produce short, warm, context-aware nudges. The project uses Unsloth and Modal for training, then deploys on a Hugging Face Space with Gradio, transformers, peft, and a runtime LoRA adapter.
This Hugging Face Blog community post introduces NeuroBait, a small AI project designed to initiate difficult tasks for ADHD. The author's starting point is not to diagnose ADHD or create another to-do list, but to observe a common dilemma among ADHD users: they often know what to do, and the real obstacle lies in the breakpoint between "knowing" and "starting." Using the partner's daily experiences as an example, the author argues that many ADHD tools are too biased toward tests, checklists, and theoretical frameworks, which may increase the choice burden when users are already frozen.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Related
Summaries are AI-generated; the original article is authoritative.