Mechanistic Interpretability Hackathon

Join us for the Mechanistic Interpretability Hackathon, inspired by similar hackathons hosted by Apart Research(link) in the past. In this hackathon, you or your team will work on a mechanistic interpretability research project during a weekend. Mechanistic interpretability is a field focused on reverse-engineering neural networks.

Event Details

• Location: our Discord interpretability server, https://discord.gg/Gv9r4b88hZ 
• Duration: 48 hours of coding, brainstorming, and breakthroughs!

How to Participate

1. Connect in the #hackathon Channel: if you’re not already a member, join our interpretability server to say hi to other participants, form teams, and stay updated in the dedicated #hackathon channel.
2. Submission: by the end of the hackathon, submit your projects to itch.io.

Submission Evaluation

We’ll use the default itch.io voting system. There's no prize, but feel free to motivate yourself with the bragging rights of being first if you want 🙂

Schedule

• Friday 17pm (UTC): kick-off event in the Discord server.
• Saturday 17pm (UTC): Project discussion
• Sunday 17pm (UTC): Ending sesion

Project Ideas

If you want inspiration for what to work on, you can look at the inspiration and project ideas section of previous apart research hackathons:

https://itch.io/jam/interpretability

https://alignmentjam.com/jam/interpretability

You can also check projects that were submitted to those for inspiration and avoiding duplicating work.

Neel is too busy these days to give talks on random hackathon like this one but you can enjoy his talk from the previous apart research hackathon and pretend he had given it for us :b.

Contact

For any queries, reach out to Victor Levoso on the #hackathon channel on Discord.

See you at the hackathon!