nm-testing
/

nvfp4_moe-e2e

8-bit precision

compressed-tensors

Model card Files Files and versions

nm-autobot commited on about 10 hours ago

Commit

2bed791

·

verified ·

1 Parent(s): 3a0ee92

Upload folder using huggingface_hub

Files changed (2) hide show

config.json +2 -2
model.safetensors +1 -1

config.json CHANGED Viewed

@@ -46,7 +46,7 @@
           "dynamic": false,
           "group_size": 16,
           "num_bits": 4,
-          "observer": "memoryless_minmax",
           "scale_dtype": "torch.float8_e4m3fn",
           "strategy": "tensor_group",
           "symmetric": true,
@@ -113,7 +113,7 @@
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
-    "version": "0.20.2a20261103"
   },
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,

           "dynamic": false,
           "group_size": 16,
           "num_bits": 4,
+          "observer": "imatrix_mse",
           "scale_dtype": "torch.float8_e4m3fn",
           "strategy": "tensor_group",
           "symmetric": true,
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
+    "version": "0.20.2.dev7+g222b8957"
   },
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:235844dee5d24fccd5a41ce6d540ff439c091d99744c3a3aac585efb8dc2e614
 size 18096813840

 version https://git-lfs.github.com/spec/v1
+oid sha256:55fe71279f5bcf08efc6fbe8903947540bccb4507e5c8f9793368ba8032a8cea
 size 18096813840