Ivan Fioravanti ᯅ (@ivanfioravanti)평가(Eval)에서 thinking tokens가 결과와 소요 시간에 큰 영향을 주며, 모델이 과도하게 생각하는 경우를 벌점화하기 위해 평가별 최대 토큰 제한이 필요할 수 있다는 제안이다.https://x.com/ivanfioravanti/status/2045134936206963142#evals #thinkingtokens #llm #benchmark #ai
Related
🧠 Beau is an AI application designed to simulate conversations with a boyfriend, focusing on daily interactions and exch...
🧠 Beau is an AI application designed to simulate conversations with a boyfriend, focusing on daily interactions and exchanges. The app enables users to engage in routine dialogue r...
What if the CIA got Sam, Dario, Peter, Zuck & Elon in the same room to train the to identify spies? ▶️ CIA Meeting Cold ...
What if the CIA got Sam, Dario, Peter, Zuck & Elon in the same room to train the to identify spies? ▶️ CIA Meeting Cold Open - SNLhttps://youtube.com/watch?v=R_8UPhJNrZ4#snl #cia #...
🤖 ‘Jensen Huang lost his mind’: how AI’s smartest man fell for the world’s oldest trickJensen Huang’s affair with a susp...
🤖 ‘Jensen Huang lost his mind’: how AI’s smartest man fell for the world’s oldest trickJensen Huang’s affair with a suspected spy is Silicon Valley’s worst nightmare come trueThe r...