Install
LessWrong is an online forum and community dedicated to improving human reasoning and decision-making. We seek to hold true beliefs and to be effective at accomplishing our goals. Each day, we aim to be less wrong about the world than the day before. A community blog devoted to refining the art of rationality
- 898articles · 90d
- 1+ mon agolatest article
- Jun 16, 2026earliest in window
- 45%with images
- 172avg words
- Science & Technology 576
- Science & Nature 378
- STEM 268
- Computers & Electronics 158
- Software Dev. 153
- People & Society 143
- Arts, Culture, Entertainment & Media 126
- Jobs & Education 77
Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
How do neural networks regress cubic polynomials? Apparently, they use a trick invented in Milan 500 years ago — LessWrong
1+ mon, 3+ week ago (1423+ words) TL;DR: Neural networks can easily learn to regress cubic polynomials of type, but how do they do that? Results show that they internally access a va…...
Theories of Deep Learning — LessWrong
2+ mon, 2+ day ago (855+ words) NOTE: These are mathematically dense frameworks which either provide a language for formalizing the field itself or aim to explain the observed phenomena. This essay is my personal understanding of the theories, after a lot of back and forth with…...
Natural Language Autoencoders are summarizers, but do they have to be? — LessWrong
2+ mon, 4+ day ago (1392+ words) TL;DR: NLAs are an interesting idea. They do reconstruct the activations well, they can kinda audit model organisms for hidden objectives. But there…...
Power Laws in NNs: A Possible Mechanism for Inductive Bias towards Sparse Representations — LessWrong
2+ mon, 2+ week ago (875+ words) This post was produced as part of the Iliad Fellowship under the mentorship of Dmitry Vaintrob. …...
Naturally learned behaviors in deep MLPs resist detection by both human and learned algorithms — LessWrong
2+ mon, 3+ week ago (840+ words) Introduction An ambitious problem in mechanistic interpretability of neural networks is finding an input for a neural network that produces a certain…...
Introduction: Gaussian Natural Latents — LessWrong
2+ mon, 3+ week ago (22+ words) Short introductory post for my research direction: Gaussian Natural Latents. I explain the motivation and give a preview of the forthcoming results. …...