fix mpt achitectures

Signed-off-by: changwangss <[email protected]>
intel · Jun 28, 2024 · c4450af · c4450af
1 parent f1dc6b8
commit c4450af
Show file tree

Hide file tree

Showing 2 changed files with 1 addition and 53 deletions.
diff --git a/intel_extension_for_transformers/transformers/modeling/modeling_auto.py b/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
@@ -845,11 +845,7 @@ def forward(self, input: torch.Tensor) -> tuple[torch.Tensor, None]:
                 or device_map == "cpu"
                 or device_map == torch.device("cpu")
             ) and model.config.model_type == "mpt":
-                config = AutoConfig.from_pretrained(
-                    os.path.join(os.path.dirname(__file__), "mosaicml_mpt-7b_config.json"),
-                    torchscript=True
-                )
-                model.config = config
+                model.config.architectures = ["MptForCausalLM"]
             model.eval()
             model_type = model.config.model_type.replace("_", "-")
 

diff --git a/intel_extension_for_transformers/transformers/modeling/mosaicml_mpt-7b_config.json b/intel_extension_for_transformers/transformers/modeling/mosaicml_mpt-7b_config.json