chore(mlx-lm): fix tie_word_embeddings for qwen2 (#566)

* chore: fix tie_word_embeddings for qwen2

* chore: default tie_word_embeddings to True

llms/mlx_lm/models/llama.py

@@ -179,8 +179,7 @@ class Model(nn.Module):
         out, cache = self.model(inputs, cache)
         return self.lm_head(out), cache
 
-    @staticmethod
-    def sanitize(weights):
+    def sanitize(self, weights):
         # Remove unused precomputed rotary freqs
         return {
             k: v for k, v in weights.items() if "self_attn.rotary_emb.inv_freq" not in k