If you told me this was a satirical Wikipedia article I would have believed you

Yeah that must be nice to have consumer protection laws where you live 🥲

My original post was actually a lived experience. Around the start of COVID I wanted to cancel my membership at Retro Fitness, but they refused unless I came in with an essay about why in person. Yup you read that right.

Fun fact: the gym was closed to guests because of COVID

Pretty easy decision

I doubt that. What evidence do you have?

Yeah good call, I’d do this too. Just watch out though because the gym will send any unpaid bills to collections. Might be worth it though to get out of the contract - just make sure to pay off the debt within 30 days.

Yes China does all of those things. Really awesome to see you’ve challenged your own flawed reality so quickly 😄

Honestly you’re not wrong. I started to proxy and honestly it feels so good. I’m so sick of spending so much money on cards.

What? He’s always consistently been the heel of the show. The writers explaining his psychopathy to make you understand is not the same thing as signing off on it. They’ve made that more clear as the seasons have progressed.

Nice 😎

Incredible. What a magical place

You realize the implication of what you wrote right? You understand what an implication is? Trying to be cute and hiding behind technicalities is pathetic and cowardly. Which honestly is par for the course for pseudo intellectual tankie trash lmao

What academic sources have you read …

Did you read your comment? It’s you. No spin needed.

Eh it’s the illusion of speed. Scaling brought enormous returns from GPT-3 -> GPT-4 but it’s been far less significant for every major release since. To compensate for this, every research lab is coming up with new ways to extract value of it of models: CoT, RL, Agent Harness etc

However, these are all hacks to make LLMs more efficient or (try) to make them more reliable. They still have significant drawbacks which will take years (probably decades) to ever get them to the point where they can reliably replace knowledge workers. China knows this and is taking a far different approach to LLM development (not a tankie fyi). Scaling is a horrible idea which will burn billions of dollars with an astronomically low chance of return.

Jfc dude get help. Seriously.

What are you saying precisely? It’s well known that LLMs have non-deterministic output (Ilya Sutskever even claims as such). Are you saying the way it goes about retrieving tokens as deterministic?

🤮

Lmao what an absolute moron you are. If you really believe that though enjoy the taste of your own medicine :)

How is it dead through? They recently released the remake to MGS3 and it uses the Fox engine (looks like it anyway) 🤔