About
Who I am
My name is Adelaide Danilov. Here I go by the nickname Senseterna. You couldn't hear it before, as it was created simultaneously with the website. So, what is this nexus about?
I'm an AI researcher, that's the main thing. My interests span Mechanistic Interpretability, AI welfare, and, occasionally non-mainstream architectures (SNN, SSM, PCN).
Why and for Whom
This website was born after the realisation that options to share results of scientific work steadily thin out: conferences are overloaded, Arxiv receives hundreds of new submissions a day, tweets are buried in the feed. And that's not to mention the situation with less technical position/opinion posts.
Thus, here, I attempt to carve some attention for several of my works. Not all - those that I believe matter more in the Grand Scheme of Things and are worth spreading info about.
Speaking of attention, whether it operates for you via neurotransmitters or QK-matrix logits, I don't discriminate. I believe the content of this website will provide utility for owners of both attention types, be it discovered while executing a task or by their own volition.
The said content
- Articles — distilled write-ups of some of my papers. At the moment of writing this "About", they are planned to be Mechanistic Interpretability related.
- Posts — less technical things which can include my opinion on things related to Artificial Intelligence: its future, relation with humans, place in the world.
Contacts
If you feel like you want to say something to me, or me to say something to you
| personal e-mail: | senseterna@gmail.com |
| institutional e-mail: | adelaide.danilov.002@student.uni.lu |
| Telegram: | @senseterna |
Shatter the mask