Added MistralLite-AWQ-32g-4b-GEMM branch

config.json

@@ -17,7 +17,7 @@
   "num_key_value_heads": 8,
   "quantization_config": {
     "bits": 4,
-    "group_size": 128,
+    "group_size": 32,
     "modules_to_not_convert": null,
     "quant_method": "awq",
     "version": "gemm",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95960bb0b19cd967e4ef0a1397ec885e6123bbd63b5934a992f253612e42dc1d
-size 4150929384
+oid sha256:051f0a043f62057d6b4e3d36de7782c7e150162f0c320f946c71b2af2ad578af
+size 4559874424