Update README.md

Author: javahedi

README.md

@@ -88,7 +88,7 @@ To scale our $1024 \times 1024$ grid across distributed nodes, we implement **1D
 
 If the global grid size is $NX \times NY$, and we launch the application with $P$ MPI Ranks, the domain is sliced horizontally:
 
-$$\text{local\_nx} = \frac{NX}{P}$$
+$\text{local\_nx} = \frac{NX}{P}$
 
 #### Example Distribution ($1024 \times 1024$ Grid across 4 Ranks)
 * **Rank 0** : Manages Global Rows $0$ to $255$ ($\text{local\_nx} = 256$)
@@ -117,4 +117,4 @@ Instead of allocating just its active rows, each rank allocates an extended buff
 N-1 │  Index N-1  │ 🏠 Last Local Row (Boundary)     │ Sent via MPI to Down         │
     ├─────────────┼──────────────────────────────────┼──────────────────────────────┤
  N  │  Index N    │ 👻 Bottom Ghost Row (Halo)       │ Received via MPI from Down   │
-    └─────────────┴──────────────────────────────────┴──────────────────────────────┘
+    └─────────────┴──────────────────────────────────┴──────────────────────────────┘