3 ms·
Possibly because those moves have the strongest signal as a winning move in the training data. Remember LLM is nothing about logic or understanding. It's about
by JoeAltmaier 3y ago
Possibly because those moves have the strongest signal as a winning move in the training data.
Remember LLM is nothing about logic or understanding. It's about correlation alone.
- zzzeek 3y agowhich means it's not "AI"