7章 モデルの制御
Understanding Catastrophic Forgetting in Language Models via Implicit Inference