Someone should make a video game that is actively harmful, not like its propaganda for dangerous opinions but just simply playing it somehow measurably deteriorates your physical/mental/emotional health like just booting it up and playing for an hour guarantees you ruin your day
@xXSaber_SwordXx I don't know who is speaking in the video, but he says "get the overhead and you get the low at the same time"
Which makes it look like you are talking about an unblockable.
👊AVISO DE NUEVA QUEDADA👊
No será en Torrejón porque el programa "La noche Abierta" para de abril a septiembre, pero nos han cedido una sala en Rivas para un nuevo evento!!!
Más información👇
sooo ahri wasn't busted enough now we're getting bugs or am i tripping ? cuz im pretty sure the super is supposed to disappear after she gets hit, and the full combo behind is just the cherry on top :D
#2xko_bugs
@OfficialBasdClb@Siewerdenlauter@sumYwastaken@Nicolai_reborn@elonmusk Depending of the chatgpt model version it also had this kind of issue, it mostly comes down to tokenization. Each company is tackling this differently so I cannot be specific but it's not just a training data thing.
https://t.co/QbtbClSly5
GGs @2XKOFR and @FlashNo47 for organizing the team battle.
To all of your team that participated I recommend to look at this website in the future 👍:
https://t.co/hNbOzv5c62
@plnm_txt
@sumYwastaken@Nicolai_reborn@elonmusk Most LLMs are trained to predict the next piece of text, not to spell things out.
So they are good at meaning, context, and writing, but weaker at character-level tasks like spelling, or counting letters.
This test only shows whoever post it doesn't know how they work.