added quantization_config

config.json

@@ -53,4 +53,13 @@
   "transformers_version": "4.16.2",
   "use_cache": true,
   "vocab_size": 96103
+},
+
+  "quantization_config": {
+    "quant_method": "awq",
+    "zero_point": true,
+    "group_size": 128,
+    "bits": 4,
+    "version": "gemm"
+  }
 }