Upload folder using huggingface_hub

config.json

@@ -88,9 +88,11 @@
           "num_bits": 8,
           "observer": null,
           "observer_kwargs": {},
+          "scale_dtype": null,
           "strategy": "token",
           "symmetric": true,
-          "type": "int"
+          "type": "int",
+          "zp_dtype": null
         },
         "output_activations": null,
         "targets": [
@@ -104,9 +106,11 @@
           "num_bits": 8,
           "observer": "minmax",
           "observer_kwargs": {},
+          "scale_dtype": null,
           "strategy": "channel",
           "symmetric": true,
-          "type": "int"
+          "type": "int",
+          "zp_dtype": null
         }
       }
     },

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3e7a8a9aa0924de09973bd0fc1b37aa4ef75b78a7c73508c788467caa9ca4b8
+oid sha256:102ff3a935bb6c6431301567344814dc0d85cfd3069705334bc745e27b178db7
 size 3819693440

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c11d03fd82edfc467006a0dff085e4ca3225f87b0032fc934e3e9bc9083cd0c
+oid sha256:22b246939898d4825ea90dd3ccc841a3a4a44519f7bdfd0df7ef820c7ba61ef9
 size 3818189840

recipe.yaml

@@ -5,8 +5,6 @@ default_stage:
       mappings:
       - - ['re:.*self_attn.q_proj', 're:.*self_attn.k_proj', 're:.*self_attn.v_proj']
         - re:.*input_layernorm
-      - - ['re:.*mamba.in_proj']
-        - re:.*input_layernorm
       - - ['re:.*shared_mlp.input_linear', 're:.*block_sparse_moe\.input_linear\.experts\.\d+']
         - re:.*post_attention_layernorm
       ignore: []