Alexei

• AlexeiMay 31, 2025, 8:34 AM

  7 points

  7

  on: Truth or Dare

  It’s much easier for me not to say anything, but my model of Duncan would prefer for me to comment.

  Overall strong upvote from me, but I’m not doing it because: what is up with the section about Zack Davis?? I’m not disputing the veracity here, because I don’t know. But it just doesn’t feel at all necessary or even useful to have it in this post. Just comes across as kind of petty. Kind if like writing The Inferno just so you can categorize and viscerally describe your enemies in hell.

• AlexeiMay 3, 2025, 3:37 PM

  10 points

  6

  on: Does trans­lat­ing a post with an LLM af­fect its rat­ing?

  I looked at your post:

  • I don’t think I ever saw it on the front page. If I did, I don’t think the title would have made it seem worth reading.

  • Reading the first few paragraphs: I’m not sure who this post is for, what problem is it addressing or why should I care in general. It’s referencing some specific chapter out of a book I never heard of.

  • It seems like a long post, so I just skip reading it.
    
    I hope the suggestions are easy to derive from the above points.

    The writing is not amazing but I didn’t get LLM vibes.

• AlexeiNov 21, 2024, 12:28 PM

  8 points

  7

  in reply to: trevor’s comment on: What are the good ra­tio­nal­ity films?

  I’d recommend Primer over this.

• AlexeiAug 25, 2024, 11:03 AM

  4 points

  0

  on: Please stop us­ing mediocre AI art in your posts

  How do I opt into the LessWrong Watercolor Aesthetic?

• AlexeiAug 20, 2024, 6:57 PM

  9 points

  12

  on: Would you benefit from, or ob­ject to, a page with LW users’ re­acts?

  Object for the stated reason.

• AlexeiAug 11, 2024, 7:56 AM

  6 points

  0

  on: Ou­trage Bonding

  I have a personal rule that I tell people when they start bringing this energy into my life: “I’m happy to listen to you for 5 minutes a day on this topic. After that I’m out.”

• AlexeiJul 8, 2024, 12:19 PM

  2 points

  0

  in reply to: rossry’s comment on: Poker is a bad game for teach­ing epistemics. Fig­gie is a bet­ter one.

  I’m interested! But I live in Portugal, so it would need to be remote.

• AlexeiApr 8, 2024, 10:53 AM

  4 points

  1

  in reply to: omark’s comment on: The Poker The­ory of Poker Night

  The former. I think you can just explain the blinds without explaining the entire poker game.

• AlexeiApr 7, 2024, 6:55 PM

  4 points

  6

  on: The Poker The­ory of Poker Night

  I like the idea and would consider doing something like that in the future. Thanks! FWIW, I found the explanation of poker completely extraneous to the main point.

• AlexeiMar 22, 2024, 7:38 PM

  10 points

  4

  in reply to: JenniferRM’s comment on: Ver­nor Vinge, who coined the term “Tech­nolog­i­cal Sin­gu­lar­ity”, dies at 79

  Vernor suggested a principle: The bad beings nearly always optimize for engagement, for pulling you ever deeper into their influence. They want to make themselves more firmly a part of your OODA loop. The good ones send you out, away from themselves in an open ended way, but better than before.

  That is profound!

• AlexeiOct 13, 2023, 12:38 PM

  2 points

  0

  on: Cir­cle of Sup­port (Oct 14th @ 10am PST)

  Oh, I should clarify that we won’t be doing Circling. We’ll just be talking.

• AlexeiSep 9, 2023, 8:58 AM

  7 points

  0

  on: Open Thread – Au­tumn 2023

  Recently I watched “The Tangle.” It’s an indie movie written and directed by the main actor from Ink, if that means anything to you. (Ink is also an indie movie, but it’s in my top 5 of all time.) Anyway, The Tangle is set in a world right after the singularity (of sorts), but where humans haven’t fully gave up control. Don’t want to spoil too much here, but I found a lot of the ideas there that were popular 5-10 years ago in the rationalist circles. Quite unexpected for an indie movie. I really enjoyed it and I think you would too.

• AlexeiJun 13, 2023, 9:01 PM

  4 points

  in reply to: Yitz’s comment on: Yitz’s Shortform

  I’d also post in the “welcome” thread.

• AlexeiJun 2, 2023, 2:38 PM

  7 points

  4

  on: Dreams of “Matho­pe­dia”

  Before building a whole website, just try this technique on some students. Whether with just paper or a quickly built web page for a few specific concepts.

• AlexeiMay 14, 2023, 4:43 PM

  2 points

  0

  in reply to: tricky_labyrinth’s comment on: Su­per-Luigi = Luigi + (Luigi—Waluigi)

  https://​​www.lesswrong.com/​​posts/​​5spBue2z2tw4JuDCx/​​steering-gpt-2-xl-by-adding-an-activation-vector

• AlexeiMay 11, 2023, 3:51 AM

  7 points

  −1

  on: How should one feel morally about us­ing chat­bots?

  I’m not using ChatGPT or any of its ilk and plan to continue to do so for the foreseeable future. Basically for the rough reasons described by OP.

  I see people make the argument that an additional subscriber doesn’t make a big difference on the margin. But as far as individual choices consumer choices go, that’s all the leverage you have!

  I think most people would agree that the eventual logical outcome of this technology is highly volatile, potentially including some very very negative outcomes in the mix. I think basic moral logic compels us not to engage with something like that. Doing otherwise is like destroying the commons but with no easy way of reparation.

  Justifying it with “it increases my productivity” seems laughable and ironic when you consider the long term consequences.

  The way I’m approaching this internally though is kind of like most vegans approach their choice, I think. It’s becoming a life choice, a moral one, and I think ultimately the right one. But I do not want to be militaristic about it. And while everyone is using ChatGPT around me I continue to love them and will do so until the end.

• AlexeiMay 1, 2023, 11:16 PM

  17 points

  9

  on: What 2025 looks like

  Interest in enrolling in CS, AI, and ML degrees goes up 5-10x from start to end of 2023.

  I’m willing to bet it will be less than 2x.

• AlexeiMar 18, 2023, 9:43 PM

  2 points

  0

  in reply to: tricky_labyrinth’s comment on: Su­per-Luigi = Luigi + (Luigi—Waluigi)

  I don’t think I define it rigorously. Maybe someone with deeper technical understanding of these models could.

  But if I had to come up with a hack somehow, you could look at the distribution of probabilities for various words as ChatGPT is predicting the next token. Presumably you’ll noticed a certain kind of probability distribution when it’s in the “Luigi” mode and another when it’s in “Waluigi” mode. Then prodding it in the right direction might be weighing more the tokens that are a lot more frequent in the Luigi mode than Waluigi.

• AlexeiMar 2, 2023, 9:44 PM

  2 points

  0

  on: AI #2

  We have no idea how to have a program detect AI-written text in a useful way.

  This approach seems very doable:

• AlexeiFeb 22, 2023, 8:46 PM

  5 points

  0

  on: Build­ing and En­ter­tain­ing Couples

  I suppose you can call me lucky, but my wife and I had about two years of doing “quality of time spent” very well. And then we switched to building a family and that’s going well too. I guess you can have it all. 😊