Submissions from lesswrong.com

		How An Algorithm Feels From Inside (2008) (lesswrong.com)
		1 point by cubefox 23 hours ago \| past \| discuss
		Autism as the Kolmogorov complexity phenotype (lesswrong.com)
		6 points by gmanner123 1 day ago \| past \| discuss
		Autism as the Kolmogorov Complexity Phenotype (lesswrong.com)
		2 points by gmanner123 3 days ago \| past \| 3 comments
		Autism as the Kolmogorov Complexity Phenotype (lesswrong.com)
		2 points by gmanner123 3 days ago \| past \| discuss
		Getting 50% (SoTA) on Arc-AGI with GPT-4o (lesswrong.com)
		2 points by pnielsen2 4 days ago \| past \| discuss
		The Prisoner's Dilemma Is Wrong: A Case for Cooperation (lesswrong.com)
		6 points by gstenger 6 days ago \| past \| 3 comments
		Priors and Prejudice (lesswrong.com)
		3 points by surprisetalk 7 days ago \| past \| discuss
		Phone Notifications Received in 30 Minutes of Class (lesswrong.com)
		5 points by sieste 13 days ago \| past \| discuss
		Coherent Extrapolated Volition (AI) (lesswrong.com)
		1 point by Bluestein 17 days ago \| past \| 1 comment
		Six (and a half) intuitions for KL divergence (lesswrong.com)
		2 points by behnamoh 17 days ago \| past
		Transformers Represent Belief State Geometry in Their Residual Stream (lesswrong.com)
		85 points by supriyo-biswas 22 days ago \| past \| 19 comments
		Notifications Received in 30 Minutes of Class (lesswrong.com)
		2 points by monort 25 days ago \| past
		Human Are Utility Monsters (2013) (lesswrong.com)
		2 points by Dracophoenix 25 days ago \| past
		A reckless introduction to Hindley-Milner type inference (lesswrong.com)
		3 points by weatherlight 26 days ago \| past
		So has AI conquered Bridge? (2022) (lesswrong.com)
		1 point by cainxinth 27 days ago \| past
		Julia Tasks 101 (lesswrong.com)
		1 point by adgjlsfhk1 27 days ago \| past
		Notifications Received in 30 Minutes of Class (lesswrong.com)
		5 points by habryka 28 days ago \| past
		Why startup founders have mood swings (and why they may have uses) – LessWrong (lesswrong.com)
		3 points by Anon84 28 days ago \| past \| 1 comment
		Training-time domain authorization could be helpful for safety (lesswrong.com)
		1 point by darosati 29 days ago \| past
		Constructability: Designing safe-by-design AI systems from scratch (lesswrong.com)
		1 point by joy_void_joy 33 days ago \| past
		AGI is easier than robotaxis (lesswrong.com)
		3 points by herecomethefuzz 35 days ago \| past \| 1 comment
		[flagged] Ex-OpenAI employee reported losing 85% of his family's net worth (lesswrong.com)
		39 points by georgehill 35 days ago \| past \| 26 comments
		Third Time: a better way to work (2022) (lesswrong.com)
		123 points by FigurativeVoid 38 days ago \| past \| 54 comments
		Teaching CS During Take-Off (lesswrong.com)
		1 point by madspindel 39 days ago \| past
		Not Pulling the Ladder Up Behind You (lesswrong.com)
		3 points by paulpauper 39 days ago \| past
		Failures in Kindness (lesswrong.com)
		1 point by bear_with_me 40 days ago \| past
		Becoming an Amateur Polyglot (lesswrong.com)
		146 points by canthandle 41 days ago \| past \| 106 comments
		Moore's Law, AI, and the pace of progress (lesswrong.com)
		1 point by pieterr 44 days ago \| past
		Not pulling up the ladder behind you (lesswrong.com)
		1 point by sieste 47 days ago \| past
		My Hour Of Memoryless Lucidity (lesswrong.com)
		2 points by optimalsolver 49 days ago \| past
		More