-
2 weeks ago | 
lesswrong.com | Katja Grace |Adam Kaufman |Knight Lee |Logan Riggs
I'm very skeptical that a one-year pause would meaningfully reduce this 4% risk. This skepticism arises partly because I doubt much productive safety research would actually happen during such a pause. In my view, effective safety research depends heavily on an active feedback loop between technological development and broader real-world applications and integration, and pausing the technology would essentially interrupt this feedback loop.
-
Jan 21, 2025 | 
lesswrong.com | Logan Riggs
Veo-2 is google's latest video-generation model. Released Dec 16th, it's quite impressive! Of course, there are still limitations (available in that previous link), especially w/ more complex movements (e.g. skateboarder & ballerina) and consistency. Then we have this, very realistic ad created by one person in ~3 weeks. Most people would not be able to tell this is AI-generated (maybe 1/100k people could tell unprompted?). It is still human-edited and uses human voices.
-
Jan 18, 2025 | 
lesswrong.com | Logan Riggs
Confusion is a felt sense; a bodily sensation you can pay attention to and notice! So here's my question for you: Think About Things by Daði Frey Youtube Link, Spotify Link Full Lyrics: Baby I can't wait to know Believe me I'll always be there so Though I know I love you I find it hard to see how you feel about me 'Cause I don't understand you Oh you are yet to learn how to speak When we first met I will never forget 'Cause even though I didn't know you yet We were bound together then and...
-
Jan 14, 2025 | 
lesswrong.com | Logan Riggs
I remember watching Youtube videos and thinking "This is the last video, I will quit after this". However, as soon as the video ends, my preferences would suddenly change to wanting to do one more! Many of us understand this false dichotomy:Quit mid-way through the videoQuit after the video ends at 3 am So I would (sometimes) stop binging videos, but only if I quit mid-way through. Then short-form videos wrecked me.
-
Dec 4, 2024 | 
lesswrong.com | Logan Riggs
This is a special post for quick takes by. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page
-
Sep 28, 2024 | 
lesswrong.com | Raymond Arnold |Ben Pace |Thomas Kwa |Logan Riggs
One year ago, many people on LessWrong received a DM asking them to choose the most important virtue of Petrov Day, with four listed options that we'd seen people argue for in previous years.
-
Mar 27, 2024 | 
lesswrong.com | Logan Riggs
Cross-posted to EA forumThere’s been a lot of discussion among safety-concerned people about whether it was bad for Anthropic to release Claude-3. I felt like I didn’t have a great picture of all the considerations here, and I felt that people were conflating many different types of arguments for why it might be bad. So I decided to try to write down an at-least-slightly-self-contained description of my overall views and reasoning here.
-
Sep 21, 2023 | 
lesswrong.com | Logan Riggs |Aidan Ewart |Scott Emmons |Charlie Steiner
This is a linkpost for Sparse Autoencoders Find Highly Interpretable Directions in Language ModelsWe use a scalable and unsupervised method called Sparse Autoencoders to find interpretable, monosemantic features in real LLMs (Pythia-70M/410M) for both residual stream and MLPs. We showcase monosemantic features, feature replacement for Indirect Object Identification (IOI), and use OpenAI's automatic interpretation protocol to demonstrate a significant improvement in interpretability.
-
Jul 28, 2023 | 
lesswrong.com | Nina Rimsky |Ethan Perez |Sam Bowman |Logan Riggs
Produced as part of the SERI ML Alignment Theory Scholars Program - Summer 2023 Cohort, under the mentorship of Evan Hubinger. I generate an activation steering vector using Anthropic's sycophancy dataset and then find that this can be used to increase or reduce performance on TruthfulQA, indicating a common direction between sycophancy on questions of opinion and untruthfulness on questions relating to common misconceptions.
-
May 23, 2023 | 
lesswrong.com | Lee Sharkey |Logan Riggs |Joseph Van Name |Arthur Conmy
When justifying my mechanistic interpretability research interests to others, I've occasionally found it useful to borrow a distinction from physics and distinguish between 'fundamental' versus 'applied' interpretability research. Fundamental interpretability research is the kind that investigates better ways to think about the structure of the function learned by neural networks. It lets us make new categories of hypotheses about neural networks.
