text stringlengths 300 320k | source stringlengths 52 154 |
|---|---|
# GPT-3 Gems
GPT-3 is captivating, and not just because of its [potential reasoning abilities](https://www.lesswrong.com/posts/L5JSMZQvkBAx9MD5A/to-what-extent-is-gpt-3-capable-of-reasoning). This post will be a living collection of my favorite experiences with the network.
Bold text is my input; square brackets con... | https://www.lesswrong.com/posts/uu8FwG5XPQ6zpFaEN/gpt-3-gems |
# Restocking
The pandemic has illustrated the resiliency benefits of keeping extras on hand. During the grocery panic, when everyone was trying to stock up on weeks' worth of groceries at the same time, people who had extra set aside didn't need to go out to a potentially dangerous environment. Similarly, by avoiding ... | https://www.lesswrong.com/posts/epJKvpWBreWYGZCNQ/restocking |
# Are parasitic worms an effective weapon against MS?
**I'm in the process of acquiring hookworms to trial within my own body.** While I'm doing that, I'm researching the effectiveness of using immune modulating parasites to treat rheumatoid arthritis and multiple sclerosis, and I'd love a little help. My reasons for ... | https://www.lesswrong.com/posts/KG6WYZko9NzJ7JWE4/are-parasitic-worms-an-effective-weapon-against-ms |
# Covid 7/23: The Second Summit
Last week: [Covid 7/16: Becoming the Mask](https://thezvi.wordpress.com/2020/07/16/covid-7-16-becoming-the-mask/)
The news continues to be cautiously good.
Positive test rates are likely peaking. Deaths are increasing slower than one would have expected, and could well peak a few week... | https://www.lesswrong.com/posts/Nj2urkWpQhSkxC9Mj/covid-7-23-the-second-summit |
# Can you get AGI from a Transformer?
***UPDATE IN 2023: I wrote this a long time ago and you should NOT assume that I still agree with all or even most of what I wrote here. I’m keeping it posted as-is for historical interest.***
Introduction
============
I want to share my thoughts about the calculations that Tran... | https://www.lesswrong.com/posts/SkcM4hwgH3AP6iqjs/can-you-get-agi-from-a-transformer |
# Aggregating forecasts
\[EDIT: I have written an update of this post [here](https://forum.effectivealtruism.org/posts/sMjcjnnpoAQCcedL2/when-pooling-forecasts-use-the-geometric-mean-of-odds)\]
_Epistemic status: confused. I also seem to have spent some time thinking about it and still went with my gut without having... | https://www.lesswrong.com/posts/mpDGNJFYzyKkg7zc2/aggregating-forecasts |
# Bottleneck Examples
This post follows my [Chains, Bottlenecks and Optimization](https://www.lesswrong.com/posts/Ze6PqJK2jnwnhcpnb/chains-bottlenecks-and-optimization). The goal is to give hypothetical **examples** of bottlenecks and non-bottlenecks (things with excess capacity), and to answer johnswentworth, who hel... | https://www.lesswrong.com/posts/Z66PxZnZJNSXzf8TL/bottleneck-examples |
# Be impatient
June, 2017: my partner Eve and I are stuck at the visa-on-arrival desk in the domestic transfer wing of Bole International Airport, Addis Ababa. The rest of the transfer passengers, all Ethiopian, are waltzing past us to form a monstrous queue at passport control. As soon as I get my precious stamp, I s... | https://www.lesswrong.com/posts/p5vDmmoD3n8ugFgxT/be-impatient |
# Longevity interventions when young
Preface for less wrong
----------------------
\*\*Skip if you just want to read the article\*\*
I'm not 100% sure this article is in tune with LW, I usually share articles related to conceptual analysis or bashing on classical mathematics and AI risk here, those fit.
This, in th... | https://www.lesswrong.com/posts/tdMLte4yQ9fszvzBy/longevity-interventions-when-young |
# Writing with GPT-3
Cross posted, as always, [from Putanumonit](https://putanumonit.com/2020/07/22/gputanumonit-3/).
* * *
[Deep in a dungeon of words](https://play.aidungeon.io/), I face the machine that will someday replace me. But for now we are friends. The post below was written by both of us; I wrote less tha... | https://www.lesswrong.com/posts/6Efhp5YpoE9kFETDj/writing-with-gpt-3 |
# No Ultimate Goal and a Small Existential Crisis
**tl;dr** There is no ultimate goal. That may seem to imply nihilism, but it doesn't. This is fine, but in novel situations, I don't know what to do, because I don't know which goal to choose, because there is no ultimate goal.
4 year olds frequently ask "why?". Often... | https://www.lesswrong.com/posts/QQTBM4HcraWgMj4eT/no-ultimate-goal-and-a-small-existential-crisis |
# Photos Before Drywall
Houses are strange: customarily, there's no documentation. Want to know where the plumbing is? Where the wires are? Where the structure is sturdy enough to support something heavy? Sorry, the walls are in the way. You can't see it, and no one wrote down where to find it.
Sure, we have workarou... | https://www.lesswrong.com/posts/dAJJPBPuK2NFqXTMB/photos-before-drywall |
# Reveal Culture
_([crossposted from malcolmocean.com](https://malcolmocean.com/2015/06/reveal-culture/); originally published June 2015)_
I have things to say about the Ask/Guess/Tell Cultures model, and an addition/amendment to propose: Reveal Culture. Shifting cultures is hard, so what you’re about to read is not ... | https://www.lesswrong.com/posts/X6zJGshe467nxoWAZ/reveal-culture |
# Using books to prime behavior
My purpose for reading has shifted from "I want to learn this" to "I want to use this to prime my behavior".
After years of consuming psychology, business management and generally "how-to" books(as opposed to "what" books), I've learnt that I will only have gotten something tangible ou... | https://www.lesswrong.com/posts/WDzLbSahckWsanyTr/using-books-to-prime-behavior |
# Construct a portfolio to profit from AI progress.
It is probably impossible to hedge against a FOOM scenario. But it seems like there are plausible 'slow' but still transformative AI scenarios we might want to hedge against.
Here is an example of a portfolio you could construct:
40% VUG (covers most larger us tech... | https://www.lesswrong.com/posts/F2wLArj9YtZR3dpPr/construct-a-portfolio-to-profit-from-ai-progress |
# Become a person who Actually Does Things
_(This is a post from a daily blogging experiment I did at [neelnanda.io](http://neelnanda.io), which I thought might also fit the tastes of LessWrong)_
There are two kinds of people in the world: People who actually do things, and people who don’t.
Now I’ve got the unneces... | https://www.lesswrong.com/posts/CuExC9ovHQusfjyCr/become-a-person-who-actually-does-things |
# Taking the first step
_(This is a post from a daily blogging experiment I did at [neelnanda.io](http://neelnanda.io/), which I thought might also fit the tastes of LessWrong. This is about my toolkit for getting started on hard things)_
Introduction
============
A lot of my previous posts have been about prioritis... | https://www.lesswrong.com/posts/cHBpsEi3wbvzxqA6r/taking-the-first-step |
# How to learn from conversations
_(This is a post from a daily blogging experiment I did at [neelnanda.io](http://neelnanda.io), which I thought might also fit the tastes of LessWrong)_
I’m currently in the middle of a [virtual conference](https://www.eaglobal.org/events/eagxvirtual2020/), so one thing heavily on my... | https://www.lesswrong.com/posts/ora9WTeJyTHE2BXdN/how-to-learn-from-conversations |
# Constraints from naturalized ethics.
Humans are just bundles of atoms, and so for them to reflect on themselves and their goals, it's plausible that there's some interesting naturalized ethics going on. This informs how we'd think about an AI intended to do human value learning. Let's frame this as a dialogue, so my... | https://www.lesswrong.com/posts/xN8MTRN7GchFJB8WJ/constraints-from-naturalized-ethics |
# A Scalable Urban Design and the Single Building City
_Crossposted from [Medium](https://medium.com/@edabsw/a-master-plan-25dfa1ea395b). Initial description of proposed infrastructure._
_Epistemic status: I've read a few books on the subject, spent a few months trying different things and working out the details, an... | https://www.lesswrong.com/posts/fnzZCq4XCeoWGPEp8/a-scalable-urban-design-and-the-single-building-city |
# How I use Google Forms in my Workflow
I use google forms* a lot and they are a key part of my workflow. I have different forms to review my day, add to-dos and track my habits and typically fill out 3 different forms per day.
This post gives examples of forms I use and why I like them.
*\[I actually use airtable f... | https://www.lesswrong.com/posts/peN7sLWk5zEie7wxt/how-i-use-google-forms-in-my-workflow |
# Jam is obsolete
Jam is a very natural way to preserve fruit. The fruit is ripe only a small fraction of the year, you'd like to be able to eat it all year long, so you cook it with enough sugar that it won't go bad for a long time. But now that we have freezers, jam is obsolete.
When you make jam, you change the ta... | https://www.lesswrong.com/posts/STaFzyajfm3XK43Hj/jam-is-obsolete |
# You Can Probably Amplify GPT3 Directly
*Epistemological Status: The idea is un-original and the implementation is the first I'm aware of. I'm reasonably confident anyone (patient) with an hour to burn could reproduce these results. You can probably amplify GPT3 directly.*
Let me cut the suspense. I can get GPT3 to ... | https://www.lesswrong.com/posts/Mzrs4MSi58ujBLbBG/you-can-probably-amplify-gpt3-directly |
# Developmental Stages of GPTs
***Epistemic Status:** I only know as much as anyone else in my reference class (I build ML models, I can grok the GPT papers, and I don't work for OpenAI or a similar lab). But I think my thesis is original.*
Related: [Gwern on GPT-3](https://www.gwern.net/newsletter/2020/05#gpt-3)
Fo... | https://www.lesswrong.com/posts/3nDR23ksSQJ98WNDm/developmental-stages-of-gpts |
# Change the world a little bit
**Working directly in the medium**
I'm in the middle of building the largest and most complex piece of software I've ever worked on. It's a stepping-stone tool for high schoolers and undergraduates to start reading scientific literature.
The basic features are all in my head. As I cod... | https://www.lesswrong.com/posts/pt7XYxnhTQdHGf7MN/change-the-world-a-little-bit |
# Generalizing the Power-Seeking Theorems
Previously: [Seeking Power is Often Provably Instrumentally Convergent in MDPs](https://www.lesswrong.com/s/7CdoznhJaLEKHwvJW/p/6DuJxY8X45Sco4bS2#Power).
*Circa 2021, the above post was revamped to supersede this one, so I recommend just reading that instead.*
* * *
Thanks... | https://www.lesswrong.com/posts/nyDnLif4cjeRe9DSv/generalizing-the-power-seeking-theorems |
# Are we in an AI overhang?
Over on [Developmental Stages of GPTs](https://www.lesswrong.com/posts/3nDR23ksSQJ98WNDm/developmental-stages-of-gpts), orthonormal mentions
> it at least reduces the chance of a [hardware overhang.](https://aiimpacts.org/hardware-overhang/)
An _overhang_ is when you have had the ability ... | https://www.lesswrong.com/posts/N6vZEnCn6A95Xn39p/are-we-in-an-ai-overhang |
# Why You Might Want a New Way to Visualize Biases
A few days ago I made a post called [A New Way to Visualize Biases](https://www.lesswrong.com/posts/MDG4jo7iFjRpn72Af/a-new-way-to-visualize-biases), and the community seems to have been generally skeptical as to the value of the diagram. I don't think the original po... | https://www.lesswrong.com/posts/a6ZJHdHywd6dQp5t5/why-you-might-want-a-new-way-to-visualize-biases |
# [updated] how does gpt2′s training corpus capture internet discussion? not well
\[**Updated** to correct my earlier claim that this doesn't affect GPT-3. Apparently it does?\]
I’m out sick today, but had enough energy to do some GPT-related fiddling around.
This time, I was curious what “internet discussions” ten... | https://www.lesswrong.com/posts/4JeAoTrAuByXGw6zm/updated-how-does-gpt2-s-training-corpus-capture-internet |
# Basic Conversational Coordination: Micro-coordination of Intention
**Follow up to:** [Something simple to try in conversations](https://musingsandroughdrafts.wordpress.com/2017/01/03/something-simple-to-try-in-conversations/)
In this essay, I want to describe a simple, but important, abstraction for how to think a... | https://www.lesswrong.com/posts/SRgsAfpaAotH6m3mK/basic-conversational-coordination-micro-coordination-of |
# The Future of Science
_(Talk given at_ [](https://www.lesswrong.com/posts/iZ3AisoiB6qKY6Ctz/sunday-jun-28-more-online-talks-by-curated-authors) [_an event on Sunday 19th of July_](https://www.lesswrong.com/posts/eK34tm6iptbxuPgtj/sunday-july-19-1pm-pdt-talks-by-raemon-ricraz-mr-hire)_. Richard Ngo is responsible for... | https://www.lesswrong.com/posts/aEZgw6rHssGRqnRFa/the-future-of-science |
# What specific dangers arise when asking GPT-N to write an Alignment Forum post?
Last year Stuart Armstrong [announced a contest](https://www.alignmentforum.org/posts/cSzaxcmeYW6z7cgtc/contest-usd1-000-for-good-questions-to-ask-to-an-oracle-ai) to come up with the best questions to ask an Oracle AI. Wei Dai [wrote](h... | https://www.lesswrong.com/posts/Et2pWrj4nWfdNAawh/what-specific-dangers-arise-when-asking-gpt-n-to-write-an |
# Billionaire Economics
I have progressive friends in my circles and so I see a lot of stuff like this:
As well as somewhat more eloquent takes [like this one](https://mkorostoff.github.io/1-pixel-wealth/?fbclid=IwAR0keGKk-wnjNBwpzcQMrQzQTjMmy2WbSqvmg92vN5CTYtTH8l... | https://www.lesswrong.com/posts/Bbmn5CgcJKTC3qfrE/billionaire-economics |
# Asset Prices Consistently Violate Efficient Market Hypothesis
Lyall Taylor points out two reasons why asset prices consistently violate the EMH:
(1) Liquidity flywheels make the market inefficient from a long-term value perspective, because shorter-term momentum dynamics exist for structural reasons decoupled from ... | https://www.lesswrong.com/posts/iGcz2SrxrkYsJf8Yi/asset-prices-consistently-violate-efficient-market |
# Delegate a Forecast
EDIT: We've stopped answering questions for now, sorry if we didn't get to your question! We're still really interested in what kinds of questions people want forecasted, feedback on how useful it is to delegate forecasts, and Elicit as a tool, so feel free to keep commenting these thoughts. We a... | https://www.lesswrong.com/posts/sJwQPRRu6g8Lpet3o/delegate-a-forecast |
# Rereading Atlas Shrugged
_This post will not attempt to avoid spoilers, and will be much more comprehensible if you've read the book or are familiar with its basics, but I also hope it'll be somewhat understandable if you haven't read the book at all; to aid with that I'll put summaries after all the names._
I firs... | https://www.lesswrong.com/posts/xiPMaYGTm2xfsB8WF/rereading-atlas-shrugged |
# Probability that other architectures will scale as well as Transformers?
GPT-1, 2, and 3 have shown impressive scaling properties. How likely is it that, in the next five years, many other architectures will also be shown to get substantially better as they get bigger? EDIT I am open to discussion of better definiti... | https://www.lesswrong.com/posts/kpK6854ArgwySuv7D/probability-that-other-architectures-will-scale-as-well-as |
# Does the lottery ticket hypothesis suggest the scaling hypothesis?
The [lottery ticket hypothesis](https://arxiv.org/abs/1803.03635), as I (vaguely) understand it, is that artificial neural networks tend to work in the following way: When the network is randomly initialized, there is a sub-network that is already de... | https://www.lesswrong.com/posts/wFJqi75y9eW8mf8TR/does-the-lottery-ticket-hypothesis-suggest-the-scaling |
# To what extent are the scaling properties of Transformer networks exceptional?
Part of the point of GPT3 is that bigger continues to be better. ([Computerphile discussion](https://www.youtube.com/watch?v=_8yVOC4ciXc&t=392s).) [A recent question](https://www.lesswrong.com/posts/kpK6854ArgwySuv7D/probability-that-othe... | https://www.lesswrong.com/posts/KyM9p6q5SELM3Ncdu/to-what-extent-are-the-scaling-properties-of-transformer-1 |
# How will internet forums like LW be able to defend against GPT-style spam?
GPT-3 seems to be skilled enough to write forum comments that aren't easy to identify as spam. While OpenAI reduces the access to it's API it will likely don't take that long till other companies develop similar API's that are more freely ava... | https://www.lesswrong.com/posts/QZ2AR6bmc4DehRHCw/how-will-internet-forums-like-lw-be-able-to-defend-against |
# What happens to variance as neural network training is scaled? What does it imply about "lottery tickets"?
Daniel Kokotajlo [asks](https://www.lesswrong.com/posts/wFJqi75y9eW8mf8TR/does-the-lottery-ticket-hypothesis-suggest-the-scaling) whether the lottery ticket hypothesis implies the scaling hypothesis.
The way I... | https://www.lesswrong.com/posts/kFm9ZMreqeNYpg8m8/what-happens-to-variance-as-neural-network-training-is |
# Tags Discussion/Talk Thread
The LessWrong dev team is hard at work creating Talk Pages/Discussion pages for tags. When they're done, every tag page will have a corresponding talk page which lets users discuss changes and improvements related to that tag.
We don't have that yet, so in the meantime, please make comme... | https://www.lesswrong.com/posts/zZZjRCiD3KdpTxrvg/tags-discussion-talk-thread |
# Wiki-Tag FAQ
*This FAQ is specifically for the tagging system. For everything else, see the* [*general FAQ*](https://www.lesswrong.com/faq)*.*
The major sections of this FAQ are:
* [General](https://www.lesswrong.com/posts/E6CF8JCQAWqqhg7ZA/tagging-faq#General)
* [How Can I Help?](How can I help with tagging?)... | https://www.lesswrong.com/posts/E6CF8JCQAWqqhg7ZA/wiki-tag-faq |
# Tagging Open Call / Discussion Thread
You’ve listened to the LessWrong team talk about our new tagging feature for *months*. First a steady drip of *we’re working on it*, then announcements of various milestones like *you can now filter Coronavirus in or out* and *anyone can create tags*. Well, now, it's an *open c... | https://www.lesswrong.com/posts/uqXQAWxLFW8WgShtk/tagging-open-call-discussion-thread |
# Predictions for GPT-N
Regarding GPT-3, there is [some discussion whether growing the model would transform it into an Oracle AI](https://www.lesswrong.com/posts/3nDR23ksSQJ98WNDm/developmental-stages-of-gpts). I looked into the actual benchmark results (Appendix H in [the paper](https://arxiv.org/abs/2005.14165v4)) ... | https://www.lesswrong.com/posts/W3DbNmuMJLWRtE5ny/predictions-for-gpt-n |
# Parameters of Privacy
In my last post, I argued that people should [probably have more meta-discussions about privacy](https://www.lessestwrong.com/posts/rz73eva3jv267Hy7B/can-you-keep-this-confidential-how-do-you-know) (rather than simply assuming everyone was on the same page about how seriously to take confidenti... | https://www.lesswrong.com/posts/afqz78RTjGB9a7MPe/parameters-of-privacy |
# Improving local governance in fragile states - practical lessons from the field
Summary –
---------
Over the past 15 years, major international donor agencies shifted their approach to local politics in fragile states. In order to build stable state-society relations in conflict-affected societies, they committed t... | https://www.lesswrong.com/posts/z7TxsQHMh9jY2NGih/improving-local-governance-in-fragile-states-practical |
# The "best predictor is malicious optimiser" problem
Suppose you are a friendly AI A and have a mysterious black box B. B outputs a sequence of bits. You want to predict the next bits that B will output. Fortunately, you have a magic Turing machine oracle O. You can give O any computable function f( Turing machine, ... | https://www.lesswrong.com/posts/ARGXciEGtuhMKtYb8/the-best-predictor-is-malicious-optimiser-problem |
# [AN #110]: Learning features from human feedback to enable reward learning
**Newsletter #110**
Alignment Newsletter is a weekly publication with recent content relevant to AI alignment around the world. Find all Alignment Newsletter **[resources here](http://rohinshah.com/alignment-newsletter/)**. In particular, yo... | https://www.lesswrong.com/posts/P6eWEMCrSjbuWwESk/an-110-learning-features-from-human-feedback-to-enable |
# Engaging Seriously with Short Timelines
It seems like transformative AI might be coming fairly soon. By transformative AI I just mean AI that will rapidly accelerate economic and technological progress. Of course, I am not ruling out a true singularity either. I am assuming such technology can be created using v... | https://www.lesswrong.com/posts/dZoXpSa3WehwqCf2m/engaging-seriously-with-short-timelines |
# What a 20-year-lead in military tech might look like
**I've spent way too much time speculating about near-future military technology. Here is a list of technologies I think are likely to be important to a conventional war in, say, 2040.**
**If you think this list shouldn't be taken seriously because I don't ha... | https://www.lesswrong.com/posts/bffJJvCC78LZjFa3Z/what-a-20-year-lead-in-military-tech-might-look-like |
# New Paper on Herd Immunity Thresholds
Previously: [On R0](https://thezvi.wordpress.com/2020/04/07/on-r0/)
[This new paper](https://www.medrxiv.org/content/10.1101/2020.07.23.20160762v1.full.pdf) suggests that herd immunity could be achieved with only about 10% infected rather than the typically suggested 60%-70%.
... | https://www.lesswrong.com/posts/i9PZF8Ek8sgBJ9R4P/new-paper-on-herd-immunity-thresholds |
# What Failure Looks Like: Distilling the Discussion
The comments under a post often contains valuable insights and additions. They are also often very long and involved, and harder to cite than posts themselves. Given this, I was motivated to try to [distill](https://distill.pub/2017/research-debt/) some comment sect... | https://www.lesswrong.com/posts/6jkGf5WEKMpMFXZp2/what-failure-looks-like-distilling-the-discussion |
# Learning the prior and generalization
_This post is a response to Paul Christiano's post “[Learning the prior](https://www.alignmentforum.org/posts/SL9mKhgdmDKXmxwE4/learning-the-prior).”_
# The generalization problem
Generally, when we train models, we often end up deploying them in situations that are distinctly... | https://www.lesswrong.com/posts/YhQr36yGkhe6x8Fyn/learning-the-prior-and-generalization |
# Attention is your scarcest resource
Like many people, I have most of my best ideas in the shower.
This is sometimes annoying: I could use more than one shower’s worth of good ideas a day, but I’d rather not end up as a shrivelled yet insightful prune. Mostly, though, shower ideas are the incentive that keeps me sme... | https://www.lesswrong.com/posts/aDtzAZf3LnwYvmBP7/attention-is-your-scarcest-resource |
# Is the work on AI alignment relevant to GPT?
I see a lot of posts go by here on AI alignment, agent foundations, and so on, and I've seen various papers from MIRI or on arXiv. I don't follow the subject in any depth, but I am noticing a striking disconnect between the concepts appearing in those discussions and rece... | https://www.lesswrong.com/posts/dPcKrfEi87Zzr7w6H/is-the-work-on-ai-alignment-relevant-to-gpt |
# Covid 7/30: Whack a Mole
Last week: [Covid 7/23: The Second Summit](https://thezvi.wordpress.com/2020/07/23/covid-7-23-the-second-summit/)
It seems clear that positive test rates, and with them new infections, have peaked in the Southern epicenters.
Unless and until something changes, forward looking risk will con... | https://www.lesswrong.com/posts/jXeocpMHg6xuMQJux/covid-7-30-whack-a-mole |
# PSA: Tagging is Awesome
I'd like to supplement the [open call for taggers](https://www.lesswrong.com/posts/uqXQAWxLFW8WgShtk/open-call-for-taggers) with a few points:
* **Tagging is a neglected cause area.** There is still a huge amount to do, and tagging makes a real difference in making LessWrong content easier... | https://www.lesswrong.com/posts/9YsAFxGwpG6MagaST/psa-tagging-is-awesome |
# What if memes are common in highly capable minds?
The meme-theoretic view of humans says: Memes are to humans as sailors are to ships in the age of sail.
If you want to predict where a ship will go, ask: Is it currently crewed by the French or the English? Is it crewed by merchants, pirates, or soldiers? These are ... | https://www.lesswrong.com/posts/6iedrXht3GpKTQWRF/what-if-memes-are-common-in-highly-capable-minds |
# Dealing with Curiosity-Stoppers
Introduction
============
Curiosity is a [virtue](https://yudkowsky.net/rational/virtues/). It
promotes epistemic honesty, ignites creativity, and improves both competence
and well-being. Multiple posts already discussed
[different types of curiosity](https://www.lesswrong.com/posts/... | https://www.lesswrong.com/posts/fQ5bKXZ69qQ5kZhpd/dealing-with-curiosity-stoppers |
# Sunday August 2, 12pm (PDT) — talks by jimrandomh, johnswenthworth, Daniel Filan, Jacobian
This Sunday at 12pm (PDT), we're running another session of "lightning talks" by curated LessWrong authors (see [here](https://www.lesswrong.com/tag/lesswrong-events) for previous weeks' transcripts).
* Each talk will be 3-... | https://www.lesswrong.com/posts/izPFcSfhd5PWujbeX/sunday-august-2-12pm-pdt-talks-by-jimrandomh-johnswenthworth |
# Free Educational and Research Resources
This is a list of free educational resources that are frequently overlooked or somewhat obscure, but still contain a large trove of information. The selection covers software, lessons, research papers, and reference material. I will skip covering online things that I think mos... | https://www.lesswrong.com/posts/jqNWc7RpMp9t2oLMp/free-educational-and-research-resources |
# Can you gain weirdness points?
A useful idea I've been looking into more lately, is "weirdness points". Basically, some good ideas are really unusual. People are often [biased against unusual ideas](https://www.lesswrong.com/posts/atcJqdhCxTZiJSxo2/talking-snakes-a-cautionary-tale). It's often seen to be easier to f... | https://www.lesswrong.com/posts/xrceax98Bv8Jdtnqx/can-you-gain-weirdness-points-1 |
# "Go west, young man!" - Preferences in (imperfect) maps
Many people are very nationalistic, putting their country above all others. Such people can be hazy about what "above all others" can mean, outside of a few clear examples - eg winning a total war totally. They're also very hazy on what is meant by "their count... | https://www.lesswrong.com/posts/pfmFe5fgEn2weJuer/go-west-young-man-preferences-in-imperfect-maps |
# Enforcing Type Distinction
The phrase "**Enforcing Type Distinction**" was mentioned and fleshed out here 12 years ago, but only appeared exactly once and that exact phrasing didn't catch on.
I think "Enforcing Type Distinction" has value as a concept (to think more clearly), as a phrase (to discuss the thing), and... | https://www.lesswrong.com/posts/amD4wdYtsAwkBmFGg/enforcing-type-distinction |
# Let Your Mind Be Not Fixed
When I was a teenager, I knew my mind was not like adults.
Sure, I was like adults in a lot of ways. I believed things. I trusted things. I knew I was right sometimes.
But they didn't change their minds. I did about once a month!
Woe was me. How could I ever be as sure and confident as ... | https://www.lesswrong.com/posts/nmJENMKrQzmGfXAHs/let-your-mind-be-not-fixed |
# Power as Easily Exploitable Opportunities
*(Talk given at* [*an event on Sunday 28th of June*](https://www.lesswrong.com/posts/iZ3AisoiB6qKY... | https://www.lesswrong.com/posts/eqov4SEYEbeFMXegR/power-as-easily-exploitable-opportunities |
# Generalized Efficient Markets in Political Power
Schelling Points
----------------
In Thomas Schelling’s [classic experiment](https://www.amazon.com/Strategy-Conflict-New-Preface-Author/dp/0674840313), we imagine trying to meet up with someone in New York City, but we haven’t specified a time or place in advance an... | https://www.lesswrong.com/posts/YMtZRGLbvdD4BGaqN/generalized-efficient-markets-in-political-power |
# On Procrastination - The art of shaping your future actions
_(This is a post from a daily blogging experiment I did at_ _[neelnanda.io](http://neelnanda.io/), which I thought might also fit the tastes of LessWrong)_
_(Also relevant: A post I made on [systems & willpower](https://www.neelnanda.io/blog/mini-blog-post... | https://www.lesswrong.com/posts/tSmpX8ZJwcuCDfRKQ/on-procrastination-the-art-of-shaping-your-future-actions |
# Online LessWrong Community Weekend,
September 11th-13th
From 11th-13th September aspiring rationalists will gather for 3 days of socialising, fun and intellectual exploration. There will be a few pre-planned activities, but most of the content will be unconference style and participant driven.
On Monday, September ... | https://www.lesswrong.com/posts/ktBPzxKRGHHmrR5Zz/online-lesswrong-community-weekend-september-11th-13th |
# Inner Alignment: Explain like I'm 12 Edition
_(This is an unofficial explanation of Inner Alignment based on the Miri paper [Risks from Learned Optimization in Advanced Machine Learning Systems](https://arxiv.org/abs/1906.01820) (which is almost identical to the [LW sequence](https://www.lesswrong.com/s/r9tYkB2a8Fp4... | https://www.lesswrong.com/posts/AHhCrJ2KpTjsCSwbt/inner-alignment-explain-like-i-m-12-edition |
# Forecasting Newsletter: July 2020.
## Highlights
- Social Science Prediction Platform [launches](https://socialscienceprediction.org/).
- Ioannidis and Taleb [discuss](https://forecasters.org/blog/2020/06/14/covid-19-ioannidis-vs-taleb/) optimal response to COVID-19.
- Report tries to [foresee](https://reliefweb... | https://www.lesswrong.com/posts/WNdtB9pHcRhn5YYHc/forecasting-newsletter-july-2020 |
# Food Spending During Covid
Our house has [shared groceries](https://www.jefftk.com/p/shared-house-setup), and in my most recent reconcile I saw that our spending is up quite a bit relative to this time last year. In 2019, April through July, our monthly spending per person was $172, while in 2020 it was $303. In the... | https://www.lesswrong.com/posts/Z8sP5xChLSspbnyuM/food-spending-during-covid |
# What are you looking for in a Less Wrong post?
My model of "What LW as a community considers a good post" is not accurate enough for my taste. Sometimes I don't understand the karma/number of comments (or lack thereof) of some posts, and it seems rude or arrogant to ask (whether it's someone else's post or one of mi... | https://www.lesswrong.com/posts/ZEz38ae84AEvJkvgH/what-are-you-looking-for-in-a-less-wrong-post |
# Agentic Language Model Memes
Related: [What specific dangers arise when asking GPT-N to write an Alignment Forum post?](https://www.lesswrong.com/posts/Et2pWrj4nWfdNAawh/what-specific-dangers-arise-when-asking-gpt-n-to-write-an)
I've been thinking about the AI safety implications of ultra powerful language models. ... | https://www.lesswrong.com/posts/9LfyRzbK3ZLQ4Fueu/agentic-language-model-memes |
# Dissolving the "Is the Efficient Market Hypothesis Dead?" Question
*(This talk was given at* [*a public online event on Sunday July 12th*](https://www.lesswrong.com/tag/lesswrong-event-transcripts)*. Alexei is responsible for the talk, David Lambert edited the transcript. *
*If you're a curated author and intereste... | https://www.lesswrong.com/posts/3TiEZzw4ikneLGp4J/dissolving-the-is-the-efficient-market-hypothesis-dead |
# Tearing down the Chesterton's Fence principle
*If you saw a fence in the middle of nowhere, and it didn't seem to be fencing anything in or out, would you feel OK about tearing it down?*
I like Chesterton's Fence. I've seen how it can completely change somebody's thinking, even when they're angry, in one sentence. ... | https://www.lesswrong.com/posts/FtCdvfkoMf7H3XkhK/tearing-down-the-chesterton-s-fence-principle |
# When Gears Go Wrong
*(This talk was given at the* [*LessWrong curated talks event on July 19*](https://www.lesswrong.com/posts/eK34tm6iptbxuPgtj/sunday-july-19-1pm-pdt-talks-by-raemon-ricraz-mr-hire)*. mr-hire is responsible for the talk, David Lambert edited the transcript.*
*If you're a curated author and interes... | https://www.lesswrong.com/posts/A2TmYuhKJ5MbdDiwa/when-gears-go-wrong |
# Engineering Experience Through Score
This post is about bowling and Super Mario Odyssey.
Bowling is a game about knocking down pins with a ball. There are ten pins, and you (sort of; keep reading) have to knock them all down ten times (each one called a "frame"). You get only two throws per frame; if you still ... | https://www.lesswrong.com/posts/Y9Yqux7iPwpvnppyS/engineering-experience-through-score |
# Burley Bee kids bike trailer review
With the coronavirus we haven't been taking buses or taxis, which had mostly limited us to walking distance with our kids. \[1\] Expanding to bicycling distance would help a lot, so we decided to get a bicycle trailer.
I did a bunch of reading on various forums, and it looked lik... | https://www.lesswrong.com/posts/kyR3g52yvfn95ZXr7/burley-bee-kids-bike-trailer-review |
# Why isn't there a LessWrong App? Is a "blog-app" sustainable/useful for a community?
I am a newbie lesswronger, so it might be an already asked question, but I can't find an answer in the FAQs. I also don't have any idea of how many resources are needed to support a blog-app, given an already developed site/communit... | https://www.lesswrong.com/posts/FYEowAsQbZsSK8pEv/why-isn-t-there-a-lesswrong-app-is-a-blog-app-sustainable |
# What are the most important papers/post/resources to read to understand more of GPT-3?
I'm way more used to thinking about weird maths or distributed algorithms or abstract philosophical problems than about concrete machine learning architectures. But based on everything I see about GPT-3, it seems a nice idea to le... | https://www.lesswrong.com/posts/m23A54nFL5yeDRukL/what-are-the-most-important-papers-post-resources-to-read-to |
# Unifying the Simulacra Definitions
Epistemic Status: Confident this is the perspective I find most useful. This is intended to both be a stand-alone post and to be the second post in the Simulacra sequence, with the first being [Simulacra Levels and their Interactions](https://thezvi.wordpress.com/2020/06/15/simulac... | https://www.lesswrong.com/posts/QdppEcbhLTZqDDtDa/unifying-the-simulacra-definitions |
# Three mental images from thinking about AGI debate & corrigibility
Here are three mental images I’ve used when sporadically struggling to understand the ideas and prospects for [AI safety via debate](https://arxiv.org/abs/1805.00899), [IDA, and related proposals](https://www.alignmentforum.org/s/EmDuGeRw749sD3GKd). ... | https://www.lesswrong.com/posts/WjY9y7r52vaNZ2WmH/three-mental-images-from-thinking-about-agi-debate-and |
# Cryonics Cost-Benefit Analysis
*cross-posted from [niplav.site](http://niplav.site/considerations_on_cryonics.html)*
> Is [cryonics](https://en.wikipedia.org/wiki/Cryonics)
worth it, and if yes, should one
[cryocrastinate](https://alcor.org/Library/html/cryocrastination.html)
(i.e. postpone signing up for cryonics ... | https://www.lesswrong.com/posts/xwJa3ieBBQYFtCmHt/cryonics-cost-benefit-analysis |
# Solving Key Alignment Problems Group
I want to map out all the open problems in Alignment; understand the key topics, how they relate and how the parts form a greater whole. I believe a small group of people who can _notice_ _when they're confused_, can make substantial conceptual progress on many of the open probl... | https://www.lesswrong.com/posts/EnnCnnjB52MDGtiYF/solving-key-alignment-problems-group |
# is gpt-3 few-shot ready for real applications?
This is a lengthy reply to [](https://tmblr.co/mWXtvJ3Sr0NPLUIFhgtxxnA)[@the-moti](https://tmblr.co/mWXtvJ3Sr0NPLUIFhgtxxnA)‘s post [here](https://the-moti.tumblr.com/post/625378774714908672/nostalgebraist-all-the-gpt-3-excitementhype-im). Creating a new post to limi... | https://www.lesswrong.com/posts/LWdpWRyWu9aXsoZpA/is-gpt-3-few-shot-ready-for-real-applications-1 |
# What is filling the hole left by religion?
Self-help books represent the next step in the evolution of our collective consciousness. It might sound like a stretch, but if we consider that religion was once how society told its story to itself, then the [recent erosion of religion](https://whyevolutionistrue.com/2017... | https://www.lesswrong.com/posts/u59y7ciE8wtyaSoac/what-is-filling-the-hole-left-by-religion |
# A sketch of 'Simulacra Levels and their Interactions'
Two sketches i made based on [Simulacra Levels and their Interactions](https://www.lesswrong.com/posts/qDmnyEMtJkE9Wrpau/simulacra-levels-and-their-interactions). These are just sketches right now, i intend to make something better looking in the future (this is... | https://www.lesswrong.com/posts/Hd8667GH7n6MJHsj8/a-sketch-of-simulacra-levels-and-their-interactions |
# RT-LAMP is the right way to scale diagnostic testing for the coronavirus
**RT-LAMP is the right way to scale diagnostic testing for the coronavirus**
_by Kevin Fischer_
_kfischer $& gmail *( com_
[RT-LAMP](https://en.wikipedia.org/wiki/Reverse_Transcription_Loop-mediated_Isothermal_Amplification) compared to [RT-... | https://www.lesswrong.com/posts/A7sMFgcX5tbJnedCi/rt-lamp-is-the-right-way-to-scale-diagnostic-testing-for-the |
# Interpretability in ML: A Broad Overview
(Reposting because I think a GreaterWrong bug on submission made this post invisible for a while last week so I'm trying again on LW.)
This blog post is an overview of ways to think about machine learning interpretability; it covers some recent research techniques as well as... | https://www.lesswrong.com/posts/57fTWCpsAyjeAimTp/interpretability-in-ml-a-broad-overview-2 |
# Property as Coordination Minimization
A friend recently noted that they were in favor of private property, but the best defense they had to link was instead a [defense of _finance_](https://putanumonit.com/2018/12/14/defense-of-finance/). So I thought I’d give it a try. In light of a distinction people often dra... | https://www.lesswrong.com/posts/Bic9wosr4baZG87MY/property-as-coordination-minimization |
# Infinite Data/Compute Arguments in Alignment
*This is a reference post. It explains a fairly standard class of arguments, and is intended to be the opposite of novel; I just want a standard explanation to link to when invoking these arguments.*
When planning or problem-solving, we focus on the hard subproblems. If ... | https://www.lesswrong.com/posts/7CJBiHYxebTmMfGs3/infinite-data-compute-arguments-in-alignment |
# My paper was signalling the whole time - Robin Hanson wins again
Robin Hanson Meme
I recently wrote a paper on how the Jordanian monarchy decided who to give water to and who to take water away from. As I near completion, I am realizing that signalling theory from Robin Hanson ... | https://www.lesswrong.com/posts/R6MoBDCYyXAWohSac/my-paper-was-signalling-the-whole-time-robin-hanson-wins |
# Tools for keeping focused
Once I realized that [my attention was even scarcer than my time](https://www.benkuhn.net/attention/), I became an anti-distraction fanatic. During my [weekly reviews](https://www.benkuhn.net/weekly/) I methodically went through my past week, figured out what had been distracting me, and tr... | https://www.lesswrong.com/posts/mXgsd5o9uuYaQKHMz/tools-for-keeping-focused |
# [AN #111]: The Circuits hypotheses for deep learning
Alignment Newsletter is a weekly publication with recent content relevant to AI alignment around the world. Find all Alignment Newsletter **[resources here](http://rohinshah.com/alignment-newsletter/)**. In particular, you can look through **[this spreadsheet](htt... | https://www.lesswrong.com/posts/5CApLZiHGkt37nRQ2/an-111-the-circuits-hypotheses-for-deep-learning |
# Measuring hardware overhang
# Measuring hardware overhang
## Summary
How can we measure a potential AI or hardware overhang? For the problem of chess, modern algorithms gained two orders of magnitude in compute (or ten years in time) compared to older versions. While it took the supercomputer "Deep Blue" to win ove... | https://www.lesswrong.com/posts/75dnjiD8kv2khe9eQ/measuring-hardware-overhang |
# Diagramming "Replacing Guilt," Part 1
I am interested communicating ideas visually. In general, I'm able to remember images for a longer time than I remember language, and the act of refining an idea into a drawing involves wrestling with the idea a little more deeply than I would with a simple summary in text. This... | https://www.lesswrong.com/posts/TXkbCSTHL57A8BoD2/diagramming-replacing-guilt-part-1 |
# Titan (the Wealthfront of active stock picking) - What's the catch?
[Titan](https://www.titanvest.com/) is a Y Combinator startup that launched in 2018 and aims to do for active investing what Wealthfront, Betterment and Vanguard have done for passive investing.
They pick a basket of 20 companies with $10B+ market ... | https://www.lesswrong.com/posts/59oPYfFJjYn3BBBwi/titan-the-wealthfront-of-active-stock-picking-what-s-the |
# Zen and Rationality: Don't Know Mind
_This is post 1/? about the intersection of my decades of LW-style rationality practice and my several years of Zen practice._
_In today's installment, I look at the Zen notion of "Don't Know Mind" in rationalist terms._
I'm a little unsure where "don't know mind" comes from. S... | https://www.lesswrong.com/posts/vXsQWi7LMm59LMt9L/zen-and-rationality-don-t-know-mind |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.