RedHatAI
/

gemma-3-12b-it-quantized.w8a8

Image-Text-to-Text

text-generation-inference

8-bit precision

compressed-tensors

Model card Files Files and versions

nm-research commited on 6 days ago

Commit

6e50a7c

·

verified ·

1 Parent(s): 4b3734b

Update README.md

Files changed (1) hide show

README.md +4 -3

README.md CHANGED Viewed

@@ -135,9 +135,10 @@ def data_collator(batch):
 recipe = [
     GPTQModifier(
         targets="Linear",
-        scheme="W8A8",
-        ignore: ["re:.*lm_head.*", "re:.*embed_tokens.*", "re:vision_tower.*", "re:multi_modal_projector.*"],
-        sequential_update: True,
     )
 ]

 recipe = [
     GPTQModifier(
         targets="Linear",
+        ignore=["re:.*lm_head.*", "re:.*embed_tokens.*", "re:vision_tower.*", "re:multi_modal_projector.*"],
+        sequential_update=True,
+        sequential_targets=["Gemma3DecoderLayer"],
+        dampening_frac=dampening_frac,
     )
 ]