this post was submitted on 18 Nov 2024
22 points (100.0% liked)

TechTakes

1436 readers
118 users here now

Big brain tech dude got yet another clueless take over at HackerNews etc? Here's the place to vent. Orange site, VC foolishness, all welcome.

This is not debate club. Unless it’s amusing debate.

For actually-good tech, you want our NotAwfulTech community

founded 1 year ago
MODERATORS
 

Need to let loose a primal scream without collecting footnotes first? Have a sneer percolating in your system but not enough time/energy to make a whole post about it? Go forth and be mid: Welcome to the Stubsack, your first port of call for learning fresh Awful you’ll near-instantly regret.

Any awful.systems sub may be subsneered in this subthread, techtakes or no.

If your sneer seems higher quality than you thought, feel free to cut’n’paste it into its own post — there’s no quota for posting and the bar really isn’t that high.

The post Xitter web has spawned soo many “esoteric” right wing freaks, but there’s no appropriate sneer-space for them. I’m talking redscare-ish, reality challenged “culture critics” who write about everything but understand nothing. I’m talking about reply-guys who make the same 6 tweets about the same 3 subjects. They’re inescapable at this point, yet I don’t see them mocked (as much as they should be)

Like, there was one dude a while back who insisted that women couldn’t be surgeons because they didn’t believe in the moon or in stars? I think each and every one of these guys is uniquely fucked up and if I can’t escape them, I would love to sneer at them.

Last week's thread

(Semi-obligatory thanks to @dgerard for starting this)

(page 2) 50 comments
sorted by: hot top controversial new old
[–] [email protected] 10 points 5 days ago (6 children)

Dude discovers that one LLM model is not entirely shit at chess, spends time and tokens proving that other models are actually also not shit at chess.

The irony? He's comparing it against Stockfish, a computer chess engine. Computers playing chess at a superhuman level is a solved problem. LLMs have now slightly approached that level.

For one, gpt-3.5-turbo-instruct rarely suggests illegal moves,

Writeup https://dynomight.net/more-chess/

HN discussion https://news.ycombinator.com/item?id=42206817

[–] [email protected] 8 points 4 days ago* (last edited 4 days ago)

Here are the results of these three models against Stockfish—a standard chess AI—on level 1, with a maximum of 0.01 seconds to make each move

I'm not a Chess person or familiar with Stockfish so take this with a grain of salt, but I found a few interesting things perusing the code / docs which I think makes useful context.

Skill Level

I assume "level" refers to Stockfish's Skill Level option.

If I mathed right, Stockfish roughly estimates Skill Level 1 to be around 1445 ELO (source). However it says "This Elo rating has been calibrated at a time control of 60s+0.6s" so it may be significantly lower here.

Skill Level affects the search depth (appears to use depth of 1 at Skill Level 1). It also enables MultiPV 4 to compute the four best principle variations and randomly pick from them (more randomly at lower skill levels).

Move Time & Hardware

This is all independent of move time. This author used a move time of 10 milliseconds (for stockfish, no mention on how much time the LLMs got). ... or at least they did if they accounted for the "Move Overhead" option defaulting to 10 milliseconds. If they left that at it's default then 10ms - 10ms = 0ms so 🤷‍♀️.

There is also no information about the hardware or number of threads they ran this one, which I feel is important information.

Evaluation Function

After the game was over, I calculated the score after each turn in “centipawns” where a pawn is worth 100 points, and ±1500 indicates a win or loss.

Stockfish's FAQ mentions that they have gone beyond centipawns for evaluating positions, because it's strong enough that material advantage is much less relevant than it used to be. I assume it doesn't really matter at level 1 with ~0 seconds to produce moves though.

Still since the author has Stockfish handy anyway, it'd be interesting to use it in it's not handicapped form to evaluate who won.

[–] [email protected] 9 points 5 days ago

@gerikson @BlueMonday1984 the only analysis of computer chess anybody needs https://youtu.be/DpXy041BIlA

load more comments (4 replies)
[–] [email protected] 10 points 5 days ago

Stack overflow now with the sponsored crypto blogspam Joining forces: How Web2 and Web3 developers can build together

I really love the byline here. "Kindest view of one another". Seething rage at the bullshittery these "web3" fuckheads keep producing certainly isn't kind for sure.

[–] [email protected] 12 points 5 days ago* (last edited 5 days ago)

a better-thought-out announcement is coming later today, but our WriteFreely instance at gibberish.awful.systems has reached a roughly production-ready state (and you can hack on its frontend by modifying the templates, pages, static, and less directories in this repo and opening a PR)! awful.systems regulars can ask for an account and I'll DM an invite link!

[–] [email protected] 14 points 6 days ago (9 children)

The mask comes off at LWN, as two editors (jake and corbet) dive in to frantically defend the honour of Justine fucking Tunney against multiple people pointing out she's a Nazi who fills her projects with racist dogwhistles

https://lwn.net/Articles/998196/

[–] [email protected] 15 points 5 days ago (1 children)

fuck me that is some awful fucking moderation. I can’t imagine being so fucking bad at this that I:

  • dole out a ban for being rude to a fascist
  • dole out a second ban because somebody in the community did some basic fucking due diligence and found out one of the accounts defending the above fascist has been just a gigantic racist piece of shit elsewhere, surprise
  • in the process of the above, I create a safe space for a fascist and her friends

but for so many of these people, somehow that’s what moderation is? fucking wild, how the fuck did we get here

[–] [email protected] 11 points 5 days ago* (last edited 5 days ago)

See, you're assuming the goal of moderation is to maintain a healthy social space online. By definition this excludes fascists. It's that old story about how to make sure your punk bar doesn't turn into a nazi punk bar. But what if instead my goal is to keep the peace in my nazi punk bar so that the normies and casuals keep filtering in and out and making me enough money that I can stay in business? Then this strategy makes more sense.

[–] [email protected] 8 points 5 days ago (1 children)
[–] [email protected] 8 points 5 days ago (1 children)

Centrists Don't Fucking Be Like This challenge not achieved yet again

https://social.kernel.org/notice/AoGpED4fw3LSGhxTLU

[–] [email protected] 5 points 5 days ago (1 children)

fwiw this link didn't jump me to a specific reply (if you meant to highlight a particular one)

[–] [email protected] 7 points 5 days ago

It didn’t scroll for me either but there’s a reply by this corbet person with a highlighted background which I assume is the one intended to be linked to

[–] [email protected] 9 points 6 days ago (3 children)

Post by Corbet the editor. "We get it: people wish that we had not highlighted work by this particular author. Had we known more about the person in question, we might have shied away from the topic. But the article is out now, it describes a bit of interesting technology, people have had their say, please let's leave it at that."

So you updated the article to reflect this right? padme.jpg

[–] [email protected] 9 points 5 days ago

so is corbet the same kind of fucker that'll complain "everything is so political nowadays"? it seems like they are

load more comments (2 replies)
[–] [email protected] 7 points 6 days ago (1 children)

@dgerard @BlueMonday1984 also, and I know this is way beside the point, update the design of your website, motherfuckers

[–] [email protected] 5 points 4 days ago

I don't run any websites, what are you coming at me for

load more comments (5 replies)
[–] [email protected] 9 points 5 days ago (1 children)
[–] [email protected] 13 points 5 days ago

When the reporter entered the confessional, AI Jesus warned, “Do not disclose personal information under any circumstances. Use this service at your own risk.

Do not worry my child, for everything you say in this hallowed chamber is between you, AI Jesus, and the army of contractors OpenAI hires to evaluate the quality of their LLM output.

[–] [email protected] 14 points 6 days ago (2 children)

most of the dedicated Niantic (Pokemon Go, Ingress) game players I know figured the company was using their positioning data and phone sensors to help make better navigational algorithms. well surprise, it’s worse than that: they’re doing a generative AI model that looks to me like it’s tuned specifically for surveillance and warfare (though Niantic is of course just saying this kind of model can be used for robots… seagull meme, “what are the robots for, fucker? why are you being so vague about who’s asking for this type of model?”)

[–] [email protected] 8 points 5 days ago

Quick, find the guys who were taping their phones to a ceiling fan and have them get to it!

Jokes aside I'm actually curious to see what happens when this one screws up. My money is on one of the Boston Dynamics dogs running in circles about 30 feet from the intended target without even establishing line of sight. They'll certainly have to test it somehow before it starts autonomously ordering drone strikes on innocent people's homes, right? Right?

[–] [email protected] 10 points 6 days ago (1 children)

Pokemon Go To The War Crimes

[–] [email protected] 12 points 5 days ago

Pokemon Go To The Hague

[–] [email protected] 6 points 5 days ago
load more comments
view more: ‹ prev next ›