Use ngram-mod and prioritize attention experts

docs/LLM/index.md

@@ -20,7 +20,7 @@ graph LR
         B[upstream llama.cpp b10121
 frank-default-small, port 7770]
         C[Ornith-1.5-35B-A3B-CRACK Q6_K
-native MTP, selected experts offloaded via override, 491520 ctx]
+ngram-mod, selected experts offloaded via override, 491520 ctx]
         A --> B --> C
     end