Fix the region for llama3 hex-llm chat completion

PiperOrigin-RevId: 681511627

Author: vertex-mg-bot

notebooks/community/model_garden/model_garden_pytorch_llama3_1_deployment.ipynb

@@ -501,8 +501,11 @@
       "source": [
         "# @title Chat completion\n",
         "\n",
+        "_region = REGION\n",
+        "REGION = TPU_DEPLOYMENT_REGION\n",
+        "\n",
         "ENDPOINT_RESOURCE_NAME = \"projects/{}/locations/{}/endpoints/{}\".format(\n",
-        "    PROJECT_ID, TPU_DEPLOYMENT_REGION, endpoints[\"hexllm_tpu\"].name\n",
+        "    PROJECT_ID, REGION, endpoints[\"hexllm_tpu\"].name\n",
         ")\n",
         "\n",
         "# @title Chat Completions Inference\n",
@@ -564,6 +567,8 @@
         ")\n",
         "print(model_response)\n",
         "\n",
+        "REGION = _region\n",
+        "\n",
         "# @markdown Click \"Show Code\" to see more details."
       ]
     },

notebooks/community/model_garden/model_garden_pytorch_llama3_2_deployment.ipynb

@@ -489,8 +489,11 @@
       "source": [
         "# @title Chat completion\n",
         "\n",
+        "_region = REGION\n",
+        "REGION = TPU_DEPLOYMENT_REGION\n",
+        "\n",
         "ENDPOINT_RESOURCE_NAME = \"projects/{}/locations/{}/endpoints/{}\".format(\n",
-        "    PROJECT_ID, TPU_DEPLOYMENT_REGION, endpoints[\"hexllm_tpu\"].name\n",
+        "    PROJECT_ID, REGION, endpoints[\"hexllm_tpu\"].name\n",
         ")\n",
         "\n",
         "# @title Chat Completions Inference\n",
@@ -552,6 +555,8 @@
         ")\n",
         "print(model_response)\n",
         "\n",
+        "REGION = _region\n",
+        "\n",
         "# @markdown Click \"Show Code\" to see more details."
       ]
     },