diff --git a/src/ainotes.cls b/src/ainotes.cls
index b41a4ae..0323a80 100644
--- a/src/ainotes.cls
+++ b/src/ainotes.cls
@@ -39,8 +39,7 @@
     showspaces = false,
     showstringspaces = true,
     showtabs = false,
-    tabsize = 3,
-    belowskip = -0.8\baselineskip
+    tabsize = 3
 }
 \lstset{style=mystyle}
 \lstset{language=Python}
diff --git a/src/statistical-and-mathematical-methods-for-ai/sections/_gradient_methods.tex b/src/statistical-and-mathematical-methods-for-ai/sections/_gradient_methods.tex
index 0f1de01..2c1bb69 100644
--- a/src/statistical-and-mathematical-methods-for-ai/sections/_gradient_methods.tex
+++ b/src/statistical-and-mathematical-methods-for-ai/sections/_gradient_methods.tex
@@ -106,7 +106,7 @@ Note: descent methods usually converge to a local minimum.
     
     \item[Backtracking procedure] \marginnote{Backtracking procedure}
         $\alpha_k$ is chose such that it respects the Wolfe condition\footnote{\url{https://en.wikipedia.org/wiki/Wolfe_conditions}}:
-        \begin{lstlisting}[mathescape=true]
+        \begin{lstlisting}[mathescape=true, belowskip = -0.8\baselineskip]
             def backtracking($\tau$, $c_1$):
                 $\alpha_k$ = 1 # Initial guess
                 while $f(x_k - \alpha_k \nabla f(\vec{x}_k))$ > $f(\vec{x}_k)$ + $c_1 \alpha_k \nabla f(\vec{x}_k)^T \nabla f(\vec{x}_k)$: