From 9be86ac7e8672007830352c4a04f155ff57848aa Mon Sep 17 00:00:00 2001
From: Tom Jobbins <TheBloke@users.noreply.huggingface.co>
Date: Thu, 9 Nov 2023 18:20:20 +0000
Subject: [PATCH] Update for Transformers AWQ support

---
 config.json | 51 +++++++++++++++++++++++++++++----------------------
 1 file changed, 29 insertions(+), 22 deletions(-)

diff --git a/config.json b/config.json
index bb21882..da097dd 100644
--- a/config.json
+++ b/config.json
@@ -1,24 +1,31 @@
 {
-    "_name_or_path": "TheBloke/Llama-2-7B-fp16",
-    "architectures": [
-        "LlamaForCausalLM"
-    ],
-    "bos_token_id": 1,
-    "eos_token_id": 2,
-    "hidden_act": "silu",
-    "hidden_size": 4096,
-    "initializer_range": 0.02,
-    "intermediate_size": 11008,
-    "max_position_embeddings": 2048,
-    "model_type": "llama",
-    "num_attention_heads": 32,
-    "num_hidden_layers": 32,
-    "pad_token_id": 0,
-    "rms_norm_eps": 1e-05,
-    "tie_word_embeddings": false,
-    "torch_dtype": "float32",
-    "transformers_version": "4.30.2",
-    "use_cache": true,
-    "vocab_size": 32000,
-    "pretraining_tp": 1
+  "_name_or_path": "TheBloke/Llama-2-7B-fp16",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "pad_token_id": 0,
+  "rms_norm_eps": 1e-05,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.30.2",
+  "use_cache": true,
+  "vocab_size": 32000,
+  "pretraining_tp": 1,
+  "quantization_config": {
+    "quant_method": "awq",
+    "zero_point": true,
+    "group_size": 128,
+    "bits": 4,
+    "version": "gemm"
+  }
 }
\ No newline at end of file