use get_compressed_attention_impl

Signed-off-by: Kyle Sayers <[email protected]>

Author: kylesayrs

src/llmcompressor/modifiers/quantization/quantization/mixin.py

@@ -243,7 +243,7 @@ def _initialize_hooks(self, model: torch.nn.Module) -> Set[RemovableHandle]:
         hooks = set()
 
         # TODO: attnq
-        # attention_impl = enable_compressed_attention(model)
+        # attention_impl = get_compressed_attention_impl()
         # hooks |= register_calibrate_attn_hooks(self, attention_impl)
 
         for module in model.modules():