Fix wrong xpu device in DistributedType.MULTI_XPU mode (#28386)

* remove elif xpu

* remove redudant code

src/transformers/training_args.py

@@ -1844,11 +1844,6 @@ class TrainingArguments:
             device = torch.device("cuda", local_rank)
             self._n_gpu = 1
             torch.cuda.set_device(device)
-        elif is_torch_xpu_available() and "ACCELERATE_USE_XPU" not in os.environ:
-            os.environ["ACCELERATE_USE_XPU"] = "true"
-            self.distributed_state = PartialState(timeout=timedelta(seconds=self.ddp_timeout))
-            device = torch.device("xpu:0")
-            self._n_gpu = 1
         elif is_sagemaker_dp_enabled():
             self.distributed_state = PartialState(_use_sagemaker_dp=True)
             self._n_gpu = 1
@@ -1877,12 +1872,6 @@ class TrainingArguments:
         elif is_sagemaker_dp_enabled() or is_sagemaker_mp_enabled():
             # Already set _n_gpu
             pass
-        elif self.distributed_state.distributed_type == DistributedType.MULTI_XPU:
-            if "ACCELERATE_USE_XPU" not in os.environ:
-                os.environ["ACCELERATE_USE_XPU"] = "true"
-            self._n_gpu = 1
-            device = torch.device("xpu:0")
-            torch.xpu.set_device(device)
         elif self.distributed_state.distributed_type == DistributedType.NO:
             if self.use_mps_device:
                 warnings.warn(