fix broken math in k-step-ahead pred

chapter_recurrent-neural-networks/sequence.md

@@ -305,7 +305,7 @@ $$
 \ldots
 $$
 
-Generally, for an observed sequence up to $x_t$, its predicted output $\hat{x}_{t+k}$ at time step $t+k$ is called the *$k$-step-ahead prediction*. Since we have observed up to $x_{604}$, its $k$-step-ahead prediction is $\hat{x}_{604+k}$.
+Generally, for an observed sequence up to $x_t$, its predicted output $\hat{x}_{t+k}$ at time step $t+k$ is called the $k$*-step-ahead prediction*. Since we have observed up to $x_{604}$, its $k$-step-ahead prediction is $\hat{x}_{604+k}$.
 In other words, we will have to use our own predictions to make multistep-ahead predictions.
 Let us see how well this goes.