DRG sadly doesn’t have persistence (spelling?) and it leads to situations where if you lose connection you don’t continue (session’s over to you, if you manage to join back then good but you’re almost guaranteed to have the spot taken away by then) the game and if the host loses connection then every other member doesn’t continue the game. I think you don’t get any rewards either in those cases? Not sure how big is their studio but they could have certainly handled some things better with the game that’s been out for a while.

That’s fair. Lol

If you want you can xpost to !hmmm@lemmy.world because it would fit. :p

Except the rare times where the host will kick you for no reason and no warning after you commited a good chunk of your time to the session. When you’re kicked the game counts that as a lost game for you and you don’t even get to block the host or anything to avoid joining their sessions in the future when the host gets the option to prevent you from joining their sessions. I wish the game had a smarter handling for problematic players and wasn’t so biased towards hosts. Happened to me only a few times but it’s still annoying to experience that.

I couldn’t make it look good so I decided to nuke this panel instead:

Beat me to it as well.

What did it say?

https://simple-bench.com/index.html I was referring to this benchmark specifically because the point of it is to benchmark the actual reasoning capabilities of LLMs:

Simple bench is the only reasoning benchmark written in natural language at which English-speaking humans (and yes, even ‘smart highschoolers’) can score 90%+, while frontier LLMs get less than 50%. It is an encapsulation of the reasoning deficit found in AI like ChatGPT.

These questions are fully private, preventing contamination, and have been vetted by PhDs from multiple domains, as well as the author - Philip, from AI Explained - who first exposed the numerous errors in the MMLU (Aug 2023). This was celebrated by, among others Andrej Karpathy.

I’m curious how it will do on the private benchmark that ai explained made. I think it was called simple bench?

I really hope for well implemented multicommunities and community discovery. Discovering communities is hard on lemmy and when you do manage to find a relatively big amount of interesting communities you still lack a way to organise them into personalised feeds.

E: Personally I’d also be really interested in post scheduling to spread out the posts across time instead of dumping it all at once and be done with it.

Exercising equipment. Jump rope, dumbels, pullup bars, rowing thingie and such.

Disgusting.

Removed by mod

Haha, how did it go?

Holy shit! You actually made it? I found it shown for a short moment in a youtube video so I looked it up and posted there. Damn…

It’s a manga so from right to left and from top to bottom.

Good to know, sometimes stuff like this just won’t federate.

He chose the “she can ruin me” option.

I forgot to mark as NSFW when posting and because I marked it after the fact there’s a good chance the NSFW mark update won’t federate. Sorry.