Comment on AGI achieved 🤖
eager_eagle@lemmy.world 1 month agowhich model is it? I have a similar answer with 3.5, but 4o replies correctly
Comment on AGI achieved 🤖
eager_eagle@lemmy.world 1 month agowhich model is it? I have a similar answer with 3.5, but 4o replies correctly
ThirdConsul@lemmy.ml 1 month ago
If you take s look at 4o leaked instruction (prompt that is “injected” at the begining of the chat), that model is clearly ordered HOW to solve this kind of problem lol
eager_eagle@lemmy.world 1 month ago
are you sure?
ThirdConsul@lemmy.ml 1 month ago
Sorry, that was Claude 3.7, not ChatGPT 4o
github.com/…/Claude_Sonnet_3.7_New.txt#L92
eager_eagle@lemmy.world 1 month ago
ah, that’s reasonable though, considering LLMs don’t really “see” characters, it’s kind of impressive this works sometimes