mgoin
/

llama2.c-stories15M-quant-pt

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

mgoin commited on Mar 26

Commit

8271dca

•

1 Parent(s): 863d525

Update recipe.yaml

Files changed (1) hide show

recipe.yaml +11 -11

recipe.yaml CHANGED Viewed

@@ -18,14 +18,14 @@ test_stage:
           weights:
             num_bits: 8
             symmetric: False
-    # SparseGPTModifier:
-    #   sparsity: 0.0
-    #   block_size: 128
-    #   sequential_update: False
-    #   quantize: True
-    #   percdamp: 0.01
-    #   mask_structure: "0:0"
-    #   targets: [
-    #     "re:model.layers.\\d*$"
-    #   ]
-    #   target_ids: ["attention_mask", "position_ids"]

           weights:
             num_bits: 8
             symmetric: False
+    SparseGPTModifier:
+      sparsity: 0.0
+      block_size: 128
+      sequential_update: False
+      quantize: True
+      percdamp: 0.01
+      mask_structure: "0:0"
+      targets: [
+        "re:model.layers.\\d*$"
+      ]
+      target_ids: ["attention_mask", "position_ids"]