nm-testing
/

w8a16_grouped_quant-e2e

compressed-tensors

Model card Files Files and versions

nm-autobot commited on about 8 hours ago

Commit

a48a635

·

verified ·

1 Parent(s): 32dfb31

Upload folder using huggingface_hub

Files changed (2) hide show

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

config.json CHANGED Viewed

@@ -28,7 +28,7 @@
           "dynamic": false,
           "group_size": 128,
           "num_bits": 8,
-          "observer": "memoryless_minmax",
           "strategy": "group",
           "symmetric": true,
           "type": "int"
@@ -46,7 +46,7 @@
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
-    "version": "0.20.2a20261103"
   },
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,

           "dynamic": false,
           "group_size": 128,
           "num_bits": 8,
+          "observer": "imatrix_mse",
           "strategy": "group",
           "symmetric": true,
           "type": "int"
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
+    "version": "0.20.2.dev7+g222b8957"
   },
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3e937f64187e41c6d4ae65878804f8531a0640f8828f442e049bab6e7a099f5
 size 1246411368

 version https://git-lfs.github.com/spec/v1
+oid sha256:ce3aa15a2d103abeb4c95143fe5272898bffd0826af0810660ea8e2d0ab6fa91
 size 1246411368