註冊並分享邀請連結,可獲得影片播放與邀請獎勵。

Tom Figliolia
@TomFigliolia
加入 July 2024
14 正在關注    43 粉絲
We’ve found an empirical law governing plasticity loss in transformer models. The surprising part: pretraining on uniform data distributions doesn’t seem to make models immune.