Refactored code

Author: dragnil1

convert-hf-to-gguf.py

@@ -1303,6 +1303,8 @@ def write_tensors(self):
             self.gguf_writer.add_tensor(new_name, data)
 @Model.register("LlamaForCausalLM")
 class DeepseekCoderModel(Model):
+    model_arch = gguf.MODEL_ARCH.LLAMA
+    
     def set_gguf_parameters(self):
         super().set_gguf_parameters()
         head_count = self.hparams["num_attention_heads"]