Anna Wang
40 followers · 16 following · 1475 views
on the atlas — 80
- Essays on Reducing Suffering5 savers
- Stochastic parrot2 savers
- Asterisk9 savers
- What is a SCIF? Explaining the rooms that protect U.S. top secrets - Washington Post2 savers
- The Mom Test: how to talk to customers and learn if your business is a good idea when everybody is lying to you12 savers
- Foresight Institute's Tech Tree Project2 savers
- 10,000 Hours with Reid Hoffman: What I Learned – Ben Casnocha6 savers
- Project Zero: From Naptime to Big Sleep: Using Large Language Models To Catch Vulnerabilities In Real-World Code1 savers
- Media Lab - kieranhealy.org1 savers
- The Great Data Integration Schlep — LessWrong3 savers
- Distillation of Neurotech and Alignment Workshop January 2023 - LessWrong4 savers
- Supporting Follow-Up Screening for Flagged Nucleic Acid Synthesis Orders - The Council on Strategic Risks1 savers
- Dario Amodei — Machines of Loving Grace50 savers
- Can LLMs generate novel research ideas?1 savers
- On becoming competitive when joining a new company | Ludwig18 savers
- Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructure4 savers
- Electric Zine Maker (a work in progress, be gentle, hug it often) by alienmelon3 savers
- Delphi method - Wikipedia2 savers
- The current state of biodefense in the US | Eukaryote Writes Blog1 savers
- Book review: Barriers to Bioweapons | Eukaryote Writes Blog1 savers
- The Germy Paradox – The empty sky: How close did we get to BW usage? | Eukaryote Writes Blog1 savers
- My techno-optimism27 savers
- Rick Searle « Utopia or Dystopia1 savers
- About – Project Hieroglyph1 savers
- Accelerando I « Utopia or Dystopia1 savers
- How Did George Soros Break the Bank of England?1 savers
- The Pygmalion Effect: Proving Them Right2 savers
- The Internet Classics Archive | Apology by Plato3 savers
- Elections 2024: Tracking AI use in global elections - Rest of World1 savers
- Optimally Allocating Compute Between Inference and Training – Epoch1 savers
- So you wanna de-bog yourself - by Adam Mastroianni24 savers
- Fix Red Clips In DaVinci Resolve In Under 2 Minutes - YouTube1 savers
- Five quick takeaways on the chip supply chain – Emerging Technology Observatory1 savers
- Film Study for Research - Jacob Steinhardt3 savers
- Statistical Modeling, Causal Inference, and Social Science3 savers
- How Do You Know When Society Is About to Fall Apart? - The New York Times5 savers
- Emerging Technology Observatory1 savers
- AI Alignment2 savers
- AGI safety career advice - EA Forum2 savers
- AI Governance Needs Technical Work - EA Forum1 savers
- Neurotechnology is Critical for AI Alignment5 savers
- Anthropic | Core Views on AI Safety: When, Why, What, and How6 savers
- The mystery of the miracle year - by Dwarkesh Patel8 savers
- How I Cured my RSI Pain2 savers
- Nintil - Aging is already solved in vitro. What comes next?3 savers
- How to Do Great Work80 savers
- What Should You Do with Your Life? Directions and Advice - Alexey Guzey73 savers
- Cultivating a state of mind where new ideas are born65 savers
- There’s no speed limit56 savers
- How to Do Great Work51 savers
- principles - Nabeel S. Qureshi41 savers
- This is Water by David Foster Wallace (Full Transcript and Audio) - Farnam Street35 savers
- The Strength of Being Misunderstood - Sam Altman32 savers
- How To Make (almost) Anything23 savers
- LLM Visualization22 savers
- Excalidraw | Hand-drawn look & feel • Collaborative • Secure20 savers
- Augmenting Long-term Memory19 savers
- [2304.03442] Generative Agents: Interactive Simulacra of Human Behavior19 savers
- A Comprehensive Mechanistic Interpretability Explainer & Glossary - Dynalist17 savers
- History of Philosophy - Summarized & Visualized17 savers
- Using Artificial Intelligence to Augment Human Intelligence15 savers
- Why most people pick the wrong career14 savers
- A Website Is A Room13 savers
- [REPOST] Epistemic Learned Helplessness | Slate Star Codex13 savers
- [2304.03442] Generative Agents: Interactive Simulacra of Human Behavior10 savers
- Architecting Discovery: A Model for How Engineers Can Help Invent Tools for Neuroscience9 savers
- Cyborgism - LessWrong8 savers
- Trailblazer List7 savers
- Is Science Stagnant? - The Atlantic7 savers
- Successful language model evals — Jason Wei6 savers
- The Goals of Neurotechnology6 savers
- The Wicked Problem Experience5 savers
- Magna Carta Scientiae5 savers
- Butterworth filter3 savers
- “Democracy-Affirming” AI Could Make Things Even Worse3 savers
- How to (seriously) read a scientific paper | Science | AAAS3 savers
- The trouble in comparing different approaches to science funding3 savers
- Product Advice for Hardware Founders : YC Startup Library | Y Combinator3 savers
- [2004.07213] Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims2 savers
- The art, science, and labor of recruiting | Khosla Ventures2 savers
highlights — 336
We believe that this recipe will still hold true where the master weights in the parameter server will be FP32 but the actual calculations will be performed in FP8 or even lower such as MX6
Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructuredue to convergence challenges with newer model architectures, everyone just simplified their training by moving back to full synchronous gradient descent
Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructuredue to the importance of fault tolerant training, publishing of methods has effectively stopped.
Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructuremost clusters will be burnt in at both high temperatures and rapidly fluctuating temperatures for a significant amount of time to ensure that all the infant mortality failures are past and have shifted into the random failure phrase.
Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructurefault tolerance is one of the most important aspects to address in scaling clusters of 100k+ GPUs towards a single workload. Nvidia is way behind Google on reliability of their AI systems
Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructurethat merely getting a hold of a dangerous agent is the least of the problems of a theoretical bioweapons program, and that small actors are unlikely to be able to effectively pull this off now.
Book review: Barriers to Bioweapons | Eukaryote Writes BlogOne interesting takeaway is that covertness has a substantial cost – forcing a program to “go underground” is a huge impediment to progress. This suggests that the Biological Weapons Convention, which has been criticized for being toothless and lacking provisions for enforcement, is actually already doing very useful work – by forcing programs to be covert at all.
Book review: Barriers to Bioweapons | Eukaryote Writes BlogUncertainty and a lack of coordination ultimately lead the program nowhere
Book review: Barriers to Bioweapons | Eukaryote Writes BlogThe book describes the Department of Energy replacing nuclear weapons parts in the late 1990s, and realizing that they no longer knew how to make a particular foam crucial to thermonuclear warheads, that their documentation for the foam’s production was insufficient, and that anyone who had done it before was long retired. They had to spend nine years and 70 million dollars inventing a substitute for a single component
Book review: Barriers to Bioweapons | Eukaryote Writes Blogbut at the low cost of ending up on every watchlist ever,
Book review: Barriers to Bioweapons | Eukaryote Writes Blogit’d be completely predictable if they gave up their evil plans right there and started volunteering in soup kitchens instead
Book review: Barriers to Bioweapons | Eukaryote Writes BlogSubtle cues in our behavior influence what other people are capable of.
The Pygmalion Effect: Proving Them RightEvery week, reflect on what things felt less efficient than they needed to be. Think for yourself how to improve these, then talk to friends, colleagues, or mentors to get additional ideas.
Film Study for Research - Jacob SteinhardtFirst try to think about whether there’s a way to modify your thought process to reliably come up with such ideas in the future.
Film Study for Research - Jacob SteinhardtIn summary, film study blogs for off-the-cuff research thinking; watch great presentations and record yourself to learn how to speak; pair program and watch programming streams; and read histories of science for long-term research decisions.
Film Study for Research - Jacob SteinhardtTo solve this problem: We’d want people to make conscious decisions about what the AIs should do when all the normal sources of authority disagree and explicitly train the AIs for the right behavior in those circumstances. Also, we’d want them to institute controls to prevent a small number of individuals from unilaterally retraining the AIs.
Project ideas: Governance during explosive technological growthOne proposal to help rectify this power imbalance is to use the framework of solidarity — the equal and just sharing of prosperity and burdens — to more equitably allocate the gains from AI. In practice, this might look like adequately compensating anyone whose actions provide data to train an AI model and redistributing the wealth AI generates for tech companies and their executives.
“Democracy-Affirming” AI Could Make Things Even WorseUnlike other countries, the U.S. lacks national data protection legislation that would regulate the inputs to AI systems. The White House’s flagship initiative for making AI companies accountable to the public, the AI Bill of Rights blueprint, is in its own words “non-binding and does not constitute U.S. government policy.”
“Democracy-Affirming” AI Could Make Things Even WorseBasically, once you’ve made a decision, you’ll need to convince others that you’re right and get them to prioritize what you need from them over the other things on their plate.
Speed as a Habit | First Round Reviewust get their input first. Don’t get your work reversed later on
Speed as a Habit | First Round ReviewThe idea isn't that we'll ask people what their interest levels and business are: we'll try to infer it from what we can observe. And then we can use this inferred information to do what we actually want to do: guess the paper relevance for a given student.
Probabilistic ProgrammingExamples here include cyborg technologies, brain-computer interfaces etc.
How Ai Fails Us - RadXChangeAs this argument runs, the problem is not the use of reward-maximizing algorithms to optimize engagement but the fact that frms are deploying these techniques to maximize proft.
How Ai Fails Us - RadXChangeMoreover, by focusing on replacing human capacities, AEAI imposes a growth ceiling that coincides with displacement
How Ai Fails Us - RadXChangeAutomation in this period did not rely on the tools that are today sometimes labeled “AI,” which are more recent origin, but was heavily premised on a robotics tradition that grew out of the previous generation of AEAI in the 1980s.
How Ai Fails Us - RadXChangeexternal security testing
FACT SHEET: Biden-Harris Administration Secures Voluntary Commitments from Leading Artificial Intelligence Companies to Manage the Risks Posed by AI | The White Houseestablish or join a forum or mechanism through which they can develop, advance, and adopt shared standards and best practices for frontier AI safety
Ensuring-Safe-Secure-and-Trustworthy-AI.pdfa) disincentivizes customers from leaving for a competitor or b) reduces/eliminates competition
Product Advice for Hardware Founders : YC Startup Library | Y CombinatorOne thing I've learned about rejections: listen to the 'no' but not the 'why'.
Product Advice for Hardware Founders : YC Startup Library | Y CombinatorIf you get a 10x increase in compute, you should make your model 3.1x times bigger and the data you train over 3.1x bigger; if you get a 100x increase in compute, you should make your model 10x bigger and your data 10x bigger.
New Scaling Laws for Large Language Models - LessWrongThe search space is huge. It's the cartesian product of all possible types of work, both known and yet to be discovered, and all possible future versions of you.
How to Do Great Workew ideas come from doing this about nontrivial things. Which may help explain why people's reaction to a new idea is often the first half of laughing: Ha!
How to Do Great WorkOne of the most powerful kinds of copying is to copy something from one field into another.
How to Do Great WorkIndeed, the features that are easiest to imitate are the most likely to be the flaws.
How to Do Great WorkThere are definitely some dangers to copying, though. One is that you'll tend to copy old things — things that were in their day at the frontier of knowledge, but no longer are.
How to Do Great WorkNor does copying necessarily make your work unoriginal. Originality is the presence of new ideas, not the absence of old ones.
How to Do Great WorkBut perhaps the worst thing schools do to you is train you to win by hacking the test. You can't do great work by doing that. You can't trick God. So stop looking for that kind of shortcut. The way to beat the system is to focus on problems and solutions that others have overlooked, not to skimp on the work itself.
How to Do Great WorkSo when you're learning about something for the first time, pay attention to things that seem wrong or missing.
How to Do Great WorkThere's a big difference between doing something you worry might be a waste of time and doing something you know for sure will be. The former is at least a bet, and possibly a better one than you think.
How to Do Great WorkThe biggest is probably time. The young have no idea how rich they are in time. The best way to turn this time to advantage is to use it in slightly frivolous ways: to learn about something you don't need to know about, just out of curiosity, or to try building something just because it would be cool, or to become freakishly good at something.
How to Do Great WorkInexperience makes them fear risk, but it's when you're young that you can afford the most
How to Do Great WorkIf you keep projects small and use flexible media, you don't have to plan as much, and your designs can evolve instea
How to Do Great WorkAnd it is more organized; it just doesn't work as well.
How to Do Great WorkAn early version of a new project will sometimes be dismissed as a toy
How to Do Great WorkBegin by trying the simplest thing that could possibly work. Surprisingly often, it does. If it doesn't, this will at least get you started.
How to Do Great WorkYou start with something small and evolve it, and the final version is both cleverer and more ambitious than anything you could have planned
How to Do Great WorkThough it sounds more responsible to begin by studying everything that's been done before, you'll learn faster and have more fun by trying stuff
How to Do Great WorkSo start lots of small things. Being prolific is underrated.
How to Do Great WorkThe best questions grow in the answering. You notice a thread protruding from the current paradigm and try pulling on it, and it just gets longer and longer. So don't require a question to be obviously big before you try answering it. You can rarely predict that. It's hard enough even to notice the thread, let alone to predict how much will unravel if you pull on it.
How to Do Great WorkBut actually the more puzzled you are, the better, so long as (a) the things you're puzzled about matter, and (b) no one else understands them either.
How to Do Great Work