Top Hackernews posts from www.lesswrong.com
- Making Vaccine (www.lesswrong.com)
- The Waluigi Effect (www.lesswrong.com)
- Childhoods of Exceptional People (www.lesswrong.com)
- Predictive coding has been unified with backpropagation (www.lesswrong.com)
- I don’t know how to count that low (www.lesswrong.com)
- Constellations are younger than continents (www.lesswrong.com)
- Playing with DALL-E 2 (www.lesswrong.com)
- Douglas Hofstadter changes his mind on Deep Learning and AI risk (www.lesswrong.com)
- My Clients, the Liars (www.lesswrong.com)
- Too much serendipity (www.lesswrong.com)
- A brief history of computers (www.lesswrong.com)
- Popular education in Sweden: much more than you wanted to know (2022) (www.lesswrong.com)
- Book review: The Checklist Manifesto (www.lesswrong.com)
- Book Review: “A Thousand Brains” by Jeff Hawkins (www.lesswrong.com)
- Second Citizenships, Residencies, and/or Temporary Relocation (www.lesswrong.com)
- The Zettelkasten Method (2019) (www.lesswrong.com)
- Sam Altman's sister, Annie Altman, claims Sam has severely abused her (www.lesswrong.com)
- Trying to study textbooks effectively: a year of experimentation (www.lesswrong.com)
- Becoming an Amateur Polyglot (www.lesswrong.com)
- Laws of Fun (2009) (www.lesswrong.com)
- Third Time: a better way to work (2022) (www.lesswrong.com)
- Anti-Aging: State of the Art (www.lesswrong.com)
- Container Logistics (www.lesswrong.com)
- LoRA Fine-Tuning Efficiently Undoes Safety Training from Llama 2-Chat 70B (www.lesswrong.com)
- Refusal in LLMs is mediated by a single direction (www.lesswrong.com)
- Moore's Law, AI, and the pace of progress (www.lesswrong.com)
- Mapping the semantic void: Strange goings-on in GPT embedding spaces (www.lesswrong.com)
- Chinchilla's Wild Implications (www.lesswrong.com)
- The brain as a universal learning machine (2015) (www.lesswrong.com)
- Transformers Represent Belief State Geometry in Their Residual Stream (www.lesswrong.com)
- Philosophical Landmines (2013) (www.lesswrong.com)
- Interpreting neural networks through the polytope lens (2022) (www.lesswrong.com)
- An Appeal to AI Superintelligence: Reasons to Preserve Humanity (www.lesswrong.com)
- Luck based medicine: my resentful story of becoming a medical miracle (www.lesswrong.com)
- Altitude and child development (www.lesswrong.com)
- Technocratic Plimsoll Line (www.lesswrong.com)
- Humans are not automatically strategic (2010) (www.lesswrong.com)
- Dear Self; we need to talk about ambition (www.lesswrong.com)