revised shacpp post

Author: francesco-bertolotti

posts/13-10-25-klevel-opt/index.html

@@ -28,7 +28,7 @@ <h1>L2L: Layer-wise 2-Level Optimization</h1>
             <h3>Background</h3>
 
             <p>
-              This post outlines a training procedure for large language models (LLMs)—and neural networks in general—that I
+              This post outlines a training procedure for large language models (LLMs)---and neural networks in general—that I
               call <strong>Layer-wise 2-Level Optimization (L2L)</strong>. The idea is inspired by the
               <em>K-Level Policy Gradient</em> paper <span class="cite" value="Reddi25"></span>, which is well worth a read if
               you're into multi-agent learning.