AdventureAI VS ConsoleQuest

Direct link to Llama3's run through [0]. Having played around extensively with Llama 2 70B, these results are extremely impressive and exciting to me. It can do really interesting things like correct itself when it uses the wrong verb. It actually gets very close to getting the game right the first time through, but it uses "search furniture" instead of specifying the bed.

Also, its ability to understand and remember the orientation of the rooms shows levels of comprehension that Llama 2 would have completely flopped on:

> SITUATION: I've searched the floor in the living room, but found nothing of interest.

> THOUGHT: Maybe I should head to the kitchenette and see if I can find any clues there.

> COMMAND: Go east

[0] https://github.com/s-macke/AdventureAI/blob/master/assets/90...

I made a pretty cool semi-idle game recently, in the spirit of Progress Quest. It's called "Console Quest" and runs from the terminal. It'd be my pleasure if anybody wants to play :)

https://github.com/donuts-are-good/consolequest

I made it just for fun one night, and found it strangely addictive all on my own while testing because I'd want to keep playing even after verifying the thing I was testing.

P.s. my high score is 356 days in game. not sure it's possible to get higher, you get wiped out quick.