change name

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

Author: kylesayrs

src/llmcompressor/transformers/sparsification/compressed_tensors_utils.py

@@ -11,7 +11,7 @@
     CompressionFormat,
     ModelCompressor,
     SparsityCompressionConfig,
-    is_module_offloaded,
+    has_offloaded_params,
     update_offload_parameter,
 )
 from loguru import logger
@@ -162,7 +162,7 @@ def patch_tied_tensors_bug(model: torch.nn.Module):
 
         if storage_ptr(input_embed.weight) == storage_ptr(output_embed.weight):
             for module in (input_embed, output_embed):
-                if not is_module_offloaded(module):
+                if not has_offloaded_params(module):
                     # create new storage ptr for onloaded weight
                     untied_data = module.weight.data.clone()
                     module.weight.data = untied_data