Fix for KeyError on Loading LLaMA (#1978)

Author: imgaojun

vllm/model_executor/models/llama.py

@@ -322,6 +322,10 @@ def load_weights(self,
                 model_name_or_path, cache_dir, load_format, revision):
             if "rotary_emb.inv_freq" in name:
                 continue
+            if "rotary_emb.cos_cached" in name:
+                continue
+            if "rotary_emb.sin_cached" in name:
+                continue
             for (param_name, weight_name, shard_id) in stacked_params_mapping:
                 if weight_name not in name:
                     continue