LESSWRONG
LW

Dennis Akar
531120
Message
Dialogue
Subscribe

MATS Research Manager

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by
Newest
No wikitag contributions to display.
2Denizhan_Akar's Shortform
3y
7
Utility Maximization = Description Length Minimization
Dennis Akar2y40

What is the exact formal difference/relation between probability distribution P, random variable X, and causal model M?

Reply
Please don't throw your mind away
Dennis Akar3y10

And you can't just get that stuff by asking your mind to tell you that stuff; you have to let your mind do its thing sometimes, without requiring a legible justification.

I need to really internalize this. I tend to "rediscover" this and forget about it the moment things become more stressful. 

Requiring that things be explicitly justified in terms of consequences forces a dichotomy: if you want to do something for reasons other than explicit justifications in terms of consequences, then you either have to lie about its explicit consequences, or you have to not do the thing. Both options are bad.

Unless one explicit justification is simply that "this is fun to do, which is Good™ in-and-of-itself, and also <insert this article here/>"?

Overall extremely good article and makes me less guilty about engaging in playful thinking (a frequent occurrence) 

Typos:

  • using repetitive that hacks aren't easy -> using repetitive hacks that aren't easy
  • expects to be given you -> expects to be given to you
Reply
Denizhan_Akar's Shortform
Dennis Akar3y60

Testing Out Shortform and Comments

I really appreciate the existence of this functionality, since when it comes to trying to do something I either overdo or don't do at all. Shortforms make it managable and still publicly shared, so I still have enough incentive to create something nice and understandable. A much much better Twitter, and allows everyone to glimpse into the "studio" of very good writers.

Oh wow I can include images like this! Very nice.

hello
Reply
Denizhan_Akar's Shortform
Dennis Akar3y10

Kind of agree. What do you think is the most value you get from this? Was wondering about whether I should post accountability-related stuff here (e.g. by editing the post) but not sure if it makes sense.

Reply
Evaluations (of new AI Safety researchers) can be noisy
Dennis Akar3y10

I have been feeling extremely impostery lately and do agree on the critical self-evaluation tendency. For the last month or so I felt entirely stuck with even the idea of an application giving me severe anxiety. Have been overcoming this slightly lately but I think this post and the conversations it caused has made em feel better. Thank you.

Reply
Denizhan_Akar's Shortform
Dennis Akar3y*10

Successful Work Analysis

Managed to end up submitting some applications to places I think I would like to work at. I think things that helped me do this (notwithstanding chance/luck) were (in descending order of perceived priority):

  • Preventing potentially "oppressive" silence by having an audiobook/chat in the background. Not necessarily white noise chat, which can feel just irritating (not usually) and I don't tend to like music due to it influencing emotions too much. It also sometimes requires too much fiddling about with playlists etc. Examples:
    • At the time it was two friends discussing how they're thinking about their lives and video making ideas.
    • Next time I did this by listening to Connor chatting about life and AI safety. 
  • "Positively" working myself out / physically tiring myself. Went out to have long walk with a friend. Though I think physical activity is indeed important, it didn't feel as impactful as the oppressive-silence-breaking.
  • The idea of being able to tell a friend, who is not that much invested in what I'm doing, that I did this and this cool thing.

These seem to somewhat lower my sense of anxiety both for applications/rejections and sending a message to people I should send a message to (e.g. job or someone I haven't replied to in a very long time).

Reply
Denizhan_Akar's Shortform
Dennis Akar3y31

This feels like a mini-twitter. Interesting

Reply
Re-Examining LayerNorm
Dennis Akar3y31

colab notebook

this interactive notebook

check out the notebook

notebook

First link is not like the others.

Reply
Org announcement: [AC]RC
Dennis Akar3y110

I propose a compromise: 1.5 hoses. This will lead to greatness.

Reply
Load More
42Gated Attention Blocks: Preliminary Progress toward Removing Attention Head Superposition
1y
4
2Denizhan_Akar's Shortform
3y
7