fix deepseek v2 quant support. refer to ggml-org#9155

Author: molamooo

src/llama.cpp

@@ -17185,7 +17185,8 @@ 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 == LLM_TN(model.arch)(LLM_TENSOR_OUTPUT, "weight")) {
             qs.has_output = true;