diff --git a/docs/en/week06/06-2.md b/docs/en/week06/06-2.md index 3c213552b..00eeb5697 100644 --- a/docs/en/week06/06-2.md +++ b/docs/en/week06/06-2.md @@ -136,7 +136,7 @@ The inputs can have multiple $x$ vectors (more than $x_1$ and $x_2$). The system Attention is increasingly important in NLP systems that use transformer architectures or other types of attention. -The weights are data independent because z is data independent. +The weights are data dependent because z is data dependent. ## [Gated Recurrent Units (GRU)](https://www.youtube.com/watch?v=ycbMGyCPzvE&t=3549s)