Extracting Steering Vectors from J space

Discussion (6 Comments)Read Original on HackerNews

lwarfield•about 1 hour ago

If the author would like, I self computed a j lens for the 27b version of the qwen model. I used it for my own exploration in this area, and can share it if you want.

schmorptron•about 1 hour ago

I'm not the author, but I would like! Is it feasible to run on the same hardware as the 27b model itself?

lwarfield•about 1 hour ago

The using a J lens is super cheap compared to inference. You basically add a single matrix multiply per layer. You probably wouldn't even notice the overhead in a good implementation.

You should even be able to create a j lens from scratch, but it might take a while. I was able to do it in a few hours on an H100. Creating the J lens is basically the equivalent of calculating a few thousand training steps for a model (256,000 backprops in my case). I've got more details in a blog post:

https://blog.lwarfield.dev/layer-scope/

I'm currently at work and can't those matrixes up until I get home. I'll update this comment with a link later.

jasonjmcghee•about 1 hour ago

Fwiw you can just Google this for a model and often someone has done it

https://huggingface.co/eyes-ml/Qwen3.8-27B_jacobian-lens

a2ff6eeb0•about 1 hour ago

This sounds like a great foundation for an adtech startup.

If you provide free chatbot services, but sell advertisers bids on which steering vectors to use to bias towards products, based on an embedding of the prompt, I bet you'd make a ton of money. For example, Coca Cola would bid on prompts about drinks, and bias towards mentioning Coke products.

I wonder if you could also use a similar method to do product placement in GenAI images and videos, and whether ad revenue would be enough to offset the price of generation. Some ad bids can go pretty high...

jasonjmcghee•29 minutes ago

Steering can degrade and bias output

Even with basic experiments I've done, it frequently introduces much more hallucination etc and allowing arbitrary steering...

Not to mention you just can't trust a model's judgement if the highest bidder chooses what it thinks

Extracting Steering Vectors from J space

⚡ Community Insights

Discussion (6 Comments)Read Original on HackerNews