Upload folder using huggingface_hub

config.json

@@ -70,7 +70,9 @@
     },
     "format": "int-quantized",
     "global_compression_ratio": null,
-    "ignore": [],
+    "ignore": [
+      "lm_head"
+    ],
     "kv_cache_scheme": null,
     "quant_method": "compressed-tensors",
     "quantization_status": "compressed"

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c53b476d8b3e1bf4ef7c78a9e42fdddbf76bb756e26bc48ca03d37dec068080
-size 940233816
+oid sha256:15b611ab3529b3899356f165b606a9b0abc4f0d7b4861ac58dbfd5fe10b8b93a
+size 1149745568

model.safetensors.index.json

@@ -1,10 +1,9 @@
 {
   "metadata": {
-    "total_size": 15937101952
+    "total_size": 16146612352
   },
   "weight_map": {
     "lm_head.weight": "model-00004-of-00004.safetensors",
-    "lm_head.weight_scale": "model-00004-of-00004.safetensors",
     "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
     "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
     "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",

recipe.yaml

@@ -1,3 +1,5 @@
 default_stage:
   default_modifiers:
-    QuantizationModifier: {scheme: INT8}
+    QuantizationModifier:
+      ignore: [lm_head]
+      scheme: INT8