when resume from peft checkpoint, the model should be trainable (#24463)

src/transformers/trainer.py

@@ -2058,7 +2058,7 @@ class Trainer:
             # If train a model using PEFT & LoRA, assume that adapter have been saved properly.
             if hasattr(model, "active_adapter") and hasattr(model, "load_adapter"):
                 if os.path.exists(resume_from_checkpoint):
-                    model.load_adapter(resume_from_checkpoint, model.active_adapter)
+                    model.load_adapter(resume_from_checkpoint, model.active_adapter, is_trainable=True)
                 else:
                     logger.warning(
                         "The intermediate checkpoints of PEFT may not be saved correctly, "