build: AutoGPTQ for meta-llama/Llama-2-7b-chat-hf: 4bits, gr128, desc_act=False

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "llama-2-7b-chat-hf-gptq",
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
   "architectures": [
     "LlamaForCausalLM"
   ],

gptq_model-4bit-128g.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:981e70a261a93caeb561ab1ece8050b367a3dd07cfe65e052cf47fdb77439bd3
-size 3896904828
+oid sha256:e86a4da7602a401317c4a49747d63a60dd6e56d4d216dd33a8e54e4309f833a8
+size 3897013783

quantize_config.json

@@ -5,6 +5,6 @@
   "desc_act": false,
   "sym": true,
   "true_sequential": true,
-  "model_name_or_path": "llama-2-7b-chat-hf-gptq",
-  "model_file_base_name": "gptq_model-4bit-128g"
+  "model_name_or_path": null,
+  "model_file_base_name": null
 }