llama : fix qs.n_attention_wv for DeepSeek-V2 (ggml-org#9156)

Author: compilade

src/llama.cpp

@@ -17902,7 +17902,8 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s
 
         // TODO: avoid hardcoded tensor names - use the TN_* constants
         if (name.find("attn_v.weight")   != std::string::npos ||
-            name.find("attn_qkv.weight") != std::string::npos) {
+            name.find("attn_qkv.weight") != std::string::npos ||
+            name.find("attn_kv_b.weight")!= std::string::npos) {
             ++qs.n_attention_wv;
         } else if (name.find("attn_k.weight")   != std::string::npos) {
             ++qs.n_attention_wk;