I suppose it's relative.

It was essentially 2 days of work using AI to help us get it all together and we were able to release it within 48 hours of the research dropping. It wouldn't have been possible at all without using AI.

There’s is likely better!

https://www.neuronpedia.org/qwen3.6-27b/jlens?shareId=cmr2kx...

It talks about how they use a tool they call a Jacobian Lens to view inside the middle layers of LLM while it's working before it commits to a word (token).

I wanted to see if I could get a version of this running on the open models and to my surprise it worked! I ran some experiments with it and build a public facing free tool anyone can use with your own prompts.

Ask the model to describe a symbol of "three curving lines of water" and you can watch "ocean", "sea", and "surf" light up a few layers deeper before it settles on "waves".

You can also edit the internal state. Insert "fire" into the middle layer of the ocean prompt and the answer shifts to something about heat.

For fun / curiosity sake, I also developed way to let the model read its own inner workspace and then decide to suppress or amplify a concept, and run the prompt again.

Interesting finding from running it across models. J-lens beats a plain logit lens on some architectures and does nothing on others, and it isn't about size. A 0.5B Qwen reads better than a 2.8B Pythia. Every Pythia I tried gained basically nothing; the Llama and Qwen models gained a lot. https://lucid.earthpilot.ai/research

This is a 48 hour old project based on emerging research and built on a small model, a small probe set on rented GPUs - but I found it genuinely exciting. The code is open.

I also included a page context "Docent" AI agent you can chat with about whatever you see to help understand what is going on.

Happy to have folks poke around and break it.

I imagine the applications for allowing models to self-reflect / edit internal states can be useful for alignment, confidence, bias detection, etc. and this tool lets you play with the early stages of that.

Comments URL: https://news.ycombinator.com/item?id=48849618

Points: 34

# Comments: 8

Doing so automatically exempts you from CA SMOG because SMOG is a county by county law. If your address is out of state, you have no county smog obligation.

I went to get my vehicle smogged and was told I didn't need to.

Discovered, mapped and tested 129 endpoints so you can fully control Substack via an agent.

https://github.com/AnthonyDavidAdams/substack-api-reference

Comments URL: https://news.ycombinator.com/item?id=48660822

Points: 4

# Comments: 0

You’d also need monitoring as the networks done reveal exact time stamps on follow / unfollow.

You could build a little toy that automates follow and unfollow and monitors and if two people ran that they could communicate.

Or they could just DM :)

What is the cost?

Comments URL: https://news.ycombinator.com/item?id=48115130

Points: 2

# Comments: 0

Vibecoded a free app that lets you just ask your laptop how to do whatever it is you want to do.

Lookout can see your screen and uses Claude to guide you and help trouble shoot any issue.

Give it to them, or use it yourself. Totally free.

Comments URL: https://news.ycombinator.com/item?id=47764651

Points: 1

# Comments: 0

“The truth doesn’t require your participation, only bullshit does.”

https://youtu.be/Q6cR7PEyzW4

I'd love feedback, edits, blurbs, reviews, etc. Accepting pull requests for edits!

Comments URL: https://news.ycombinator.com/item?id=47404291

Points: 1

# Comments: 0