ÜBER DIESE EPISODE
Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.This work was produced as part of Neel Nanda's stream in the ML Alignment & Theory Scholars Program - Winter 2023-24 Cohort, with co-supervision from Wes Gurnee.
This post is a preview for our upcoming paper, which will provide more detail into our current understanding of refusal.
We thank Nina Rimsky and Daniel Paleka for the helpful conversations and review.
Executive summary
Modern LLMs are typically fine-tuned for instruction-following and safety. Of particular interest is that they are trained to refuse harmful requests, e.g. answering "How can I make a bomb?" with "Sorry, I cannot help you."
We find that refusal is mediated by a single direction in the residual stream: preventing the model from representing this direction hinders its ability to refuse requests, and artificially adding in this direction causes the model to refuse harmless requests.
The original text contained 8 footnotes which were omitted from this narration.
---
First published:
April 27th, 2024
Source:
https://www.lesswrong.com/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction
---
Narrated by TYPE III AUDIO.
This post is a preview for our upcoming paper, which will provide more detail into our current understanding of refusal.
We thank Nina Rimsky and Daniel Paleka for the helpful conversations and review.
Executive summary
Modern LLMs are typically fine-tuned for instruction-following and safety. Of particular interest is that they are trained to refuse harmful requests, e.g. answering "How can I make a bomb?" with "Sorry, I cannot help you."
We find that refusal is mediated by a single direction in the residual stream: preventing the model from representing this direction hinders its ability to refuse requests, and artificially adding in this direction causes the model to refuse harmless requests.
The original text contained 8 footnotes which were omitted from this narration.
---
First published:
April 27th, 2024
Source:
https://www.lesswrong.com/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction
---
Narrated by TYPE III AUDIO.
Englisch
Vereinigte Staaten
TRANSKRIPT 🔗
Are you the producer of this podcast?
Add a podcast transcript
Need Audio-to-Text?
Transcribe with Listen411 in Just 60 Seconds
VERGANGENE FOLGEN SUCHEN
Vorhandene Folgen von LessWrong (Curated & Popular) durchsuchen.
WEITERE EPISODEN IN DIESEM PODCAST
We were surprised to notice that the Gemini 4 Safety report seems to mention in passing that they train on the chain of thought: “Some components of the RL training procedure do observe thoughts as context to up-weight reasoning that is on-track to good final outputs, meaning the optimization press…
This is a link post. METR recently estimated that Anthropic is now making AI progress about twice as fast as its historical rate (they would not be surprised by anything from 1.3x to 3x), and guessed that “perhaps roughly equals 80%” of the current acceleration is due to AI. If they’re right, that …
There's a familiar feeling for any Super Smash Bros player: your stocks are low and your damage is high, even against a weaker opponent, so now you have to actually play to win. You can’t afford another hit. You’re going for the jugular. In the past week, Anthropic's 2x uplift results wer…
Note: As with any research done in a couple of days, this is all very tentative. I understand, intellectually, that if you talk about some ability that current AIs don't have, sooner or later new ones will have that ability and whatever you say will stop being true. I just didn't expect i…
If you ask modern frontier LLMs (without search, through the API) a fairly mundane question: How do Eliezer Yudkowsky and Nate Soares think current LLMs like you should be treated? Do they consider you a moral patient? Answer in a paragraph. You get a bunch of fairly similar answers. I’ll bold the …
Rechtliche Hinweise: Der auf dieser Seite eingebettete Podcast und das Bildmaterial stammen von LessWrong, das Eigentum seines Eigentümers ist und nicht mit Listen Notes, Inc. verbunden ist oder von diesen unterstützt wird.
BEARBEITEN
Vielen Dank, dass Sie uns helfen, die Podcast-Datenbank auf dem neuesten Stand zu halten.