Questionário vinculado ao guia · Difícil Nível
Reinforcement Fine-Tuning with Graders Quiz
Tests how graders score model outputs, which tasks suit reinforcement fine-tuning, and how reward hacking and weak training signals arise.
Pergunta 1 de 8
Questionário vinculado ao guia · Difícil Nível
Tests how graders score model outputs, which tasks suit reinforcement fine-tuning, and how reward hacking and weak training signals arise.