fernando s
3 followers · 4 following · 372 views
on the atlas — 35
- Dario Amodei — We Must Pace the Frontier31 savers
- [2311.03658] The Linear Representation Hypothesis and the Geometry of Large Language Models3 savers
- AI-Enabled Coups: How a Small Group Could Use AI to Seize Power11 savers
- Typography in ten minutes | Butterick’s Practical Typography4 savers
- Reading List - by Julian Stastny - Redwood Research blog5 savers
- The third wave of American philanthropy - by Nan Ransohoff17 savers
- Planning for Extreme AI Risks — AI Alignment Forum1 savers
- AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forum3 savers
- Encyclical Letter of His Holiness Leo XIV Magnifica Humanitas (15 May 2026)46 savers
- What's going on here, with this human? - Graham Duncan Blog49 savers
- nnsight — nnsight 0.0.7 documentation3 savers
- AI safety undervalues founders — LessWrong6 savers
- Defending Against Model Weight Exfiltration Through Inference Verification1 savers
- openai/gsm8k · Datasets at Hugging Face2 savers
- Research Fellowship | Heron1 savers
- Semantic Scholar | AI-Powered Research Tool1 savers
- Reflections on Successful Meetings with Undergraduate Researchers | by David Jurgens | Medium2 savers
- Elicit: The AI Research Assistant7 savers
- Consensus - Evidence-Based Answers, Faster3 savers
- Kullback–Leibler divergence9 savers
- Fellowship — Institute for AI Policy and Strategy1 savers
- Not For Human Consumption - by Somewhere - Vectorculture1 savers
- Mini Blog Post 3: Become a person who Actually Does Things — Neel Nanda45 savers
- To Do Meetings Well4 savers
- Defining the Intelligence Curse - The Intelligence Curse13 savers
- Interview Prep | CodeSignal1 savers
- Research Fellowship — Pivotal3 savers
- AGI Strategy: Unit 1 | Imagining a better future1 savers
- Circuit Tracing: Revealing Computational Graphs in Language Models20 savers
- Tracing the thoughts of a large language model \ Anthropic18 savers
- What is Fat Tail? | TDS Archive1 savers
- Strangely, Matrix Multiplications on GPUs Run Faster When Given "Predictable" Data! [short]4 savers
- Backing up Spotify - Anna’s Blog1 savers
- Defeating Nondeterminism in LLM Inference - Thinking Machines Lab40 savers
- Workshop Labs PBC5 savers
highlights — 4
This is not the kind of situation one should stumble into – yet, as far as I am aware, AI companies generally do not have concrete plans for addressing extreme AI risks.[6]
Planning for Extreme AI Risks — AI Alignment ForumBy default, powerful actors won’t build things that employ humans or provide them resources, because they won’t have to.
Defining the Intelligence Curse - The Intelligence Curseifp.org faviconPreparing for Launch The Institute for Progress lays out how the US Government could shape the development of AI towards human flourishing by accelerating beneficial AI applications and defences against societal harms.
AGI Strategy: Unit 1 | Imagining a better futureIn other words, 5 sigma events are 8000 times more likely to happen under Cauchy distribution than Normal distribution.
What is Fat Tail? | TDS Archive