minus-squarerecursiveInsurgent@lemm.eetoProgramming@programming.dev•Anthropic can now track the bizarre inner workings of a large language modellinkfedilinkEnglisharrow-up4arrow-down2·12 days agoInteresting how these findings refute the assertion that LLMs are just predicting the next word. Sometimes they plan ahead. linkfedilink
minus-squarerecursiveInsurgent@lemm.eetoProgramming@programming.dev•SQL NoirlinkfedilinkEnglisharrow-up3·1 month agoThis seems like it was made specifically for me and my interests. linkfedilink
Interesting how these findings refute the assertion that LLMs are just predicting the next word. Sometimes they plan ahead.