Update README.md

README.md

@@ -30,7 +30,7 @@ $O(L)$ complexity in sequence length $L$, instead of the $O(L^2)$ complexity of
 | | |
 |---|---|
 | Layers | 12 |
-| Hidden size ($d_{\text{model}}$) | 512 |
+| Hidden size  | 512 |
 | State dimension | 16 |
 | Parameters | ~105M |
 | Max sequence length | 512 |