update model.

config.json

@@ -15,6 +15,19 @@
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
   "num_key_value_heads": 8,
+  "quantization_config": {
+    "bits": 4,
+    "damp_percent": 0.01,
+    "desc_act": false,
+    "group_size": 128,
+    "is_marlin_format": false,
+    "model_file_base_name": "model",
+    "model_name_or_path": null,
+    "quant_method": "gptq",
+    "static_groups": false,
+    "sym": false,
+    "true_sequential": false
+  },
   "rms_norm_eps": 1e-05,
   "rope_theta": 10000.0,
   "sliding_window": 4096,

quantize_config.json

@@ -7,7 +7,7 @@
   "sym": false,
   "true_sequential": false,
   "model_name_or_path": null,
-  "model_file_base_name": "autoround-model-4bit-128g",
+  "model_file_base_name": "model",
   "is_marlin_format": false,
   "quant_method": "intel/auto-round"
 }