Thom_Wolf· @Thom_Wolf · X·· 4 天前AI 评分
Karpathy 离开X,Ben Affleck 讲微调
Karpathy:从 X 上消失 Ben Affleck:好了,都过来,所以你们要先冻结 base weights,学习率 2e-4
Karpathy: disappears from X
Ben Affleck: alright, gather round, so you'll want to freeze the base weights first, learning rate 2e-4
来源:Thom_Wolf · x.com