Upload folder using huggingface_hub

config.json

@@ -32,14 +32,14 @@
             "num_bits": 8,
             "observer": "minmax",
             "observer_kwargs": {},
-            "strategy": "channel",
+            "strategy": "tensor",
             "symmetric": true,
             "type": "int"
           }
         }
       },
       "format": "int-quantized",
-      "global_compression_ratio": 1.2390773684863086,
+      "global_compression_ratio": 1.2391304140415598,
       "ignore": [
         "lm_head"
       ],
@@ -48,7 +48,7 @@
     },
     "sparsity_config": {
       "format": "dense",
-      "global_sparsity": 1.188444907528722,
+      "global_sparsity": 7.819650685473849,
       "registry_requires_subclass": false,
       "sparsity_structure": "unstructured"
     }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed7796316a087eaf71963e35ff79ae097d02000411fc834b315d0bf399ad870c
-size 1232041608
+oid sha256:68c532b04501830ec5a92b14272c2cb59b8b419a2c521e57152457fb88243d5c
+size 1231252556

recipe.yaml

@@ -5,6 +5,6 @@ quant_stage:
       ignore: [lm_head]
       config_groups:
         group_0:
-          weights: {num_bits: 8, type: int, symmetric: true, strategy: channel}
+          weights: {num_bits: 8, type: int, symmetric: true, strategy: tensor}
           input_activations: {num_bits: 8, type: int, symmetric: true, dynamic: true, strategy: token}
           targets: [Linear]