Skip to content
onsuggest
Log in
Article en

Gradient Routing Creates a Headstart

Source: LessWrong ·

Tl;dr GRAFT showed that we can make reward hacking removable. In the setup we study, a small LLM has two trainable adapters, each consisting of an additional MLP in every layer. Training updates from detected hacks are routed only into one, called FORGET, which is removed ('forgotten') after training to suppress the behaviour. We reproduced a small-model version of GRAFT and tested what makes this isolation work, where hacking concentrates across layers, and the effect of adapter width. In order to conclude that these results are general, we’d need to run larger models in different settings. W

What is Gradient Routing Creates a Headstart?

Gradient Routing Creates a Headstart is a article discovery curated from LessWrong. Tl;dr GRAFT showed that we can make reward hacking removable. In the setup we study, a small LLM has two trainable adapters, each consisting of an additional MLP in every layer…

How can I access or experience this?

You can experience this content directly on its original host (LessWrong) by clicking the "Open Original" button.

Where does this content come from?

Curated from LessWrong by the OnSuggest team to surface the most interesting and serendipitous content on the web.

Keyboard shortcuts

Previous discovery
←P
Next discovery
→N
Like
L
Not for me
X
Save
S
Show this help
?
Close
Esc

Shortcuts work on the discovery page.