continue to fix distributed_type from TPU to XLA in LM examples (#38652)

examples/pytorch/image-pretraining/run_mim_no_trainer.py

@@ -625,7 +625,7 @@ def main():
     )
 
     # On TPU, the tie weights in our model have been disconnected, so we need to restore the ties.
-    if accelerator.distributed_type == DistributedType.TPU:
+    if accelerator.distributed_type == DistributedType.XLA:
         model.tie_weights()
 
     # We need to recalculate our total training steps as the size of the training dataloader may have changed.

examples/pytorch/language-modeling/run_clm_no_trainer.py

@@ -531,7 +531,7 @@ def main():
     )
 
     # On TPU, the tie weights in our model have been disconnected, so we need to restore the ties.
-    if accelerator.distributed_type == DistributedType.TPU:
+    if accelerator.distributed_type == DistributedType.XLA:
         model.tie_weights()
 
     # We need to recalculate our total training steps as the size of the training dataloader may have changed.

examples/pytorch/language-modeling/run_fim_no_trainer.py

@@ -729,7 +729,7 @@ def main():
     )
 
     # On TPU, the tie weights in our model have been disconnected, so we need to restore the ties.
-    if accelerator.distributed_type == DistributedType.TPU:
+    if accelerator.distributed_type == DistributedType.XLA:
         model.tie_weights()
 
     # We need to recalculate our total training steps as the size of the training dataloader may have changed.

examples/pytorch/language-modeling/run_mlm_no_trainer.py

@@ -568,7 +568,7 @@ def main():
     )
 
     # On TPU, the tie weights in our model have been disconnected, so we need to restore the ties.
-    if accelerator.distributed_type == DistributedType.TPU:
+    if accelerator.distributed_type == DistributedType.XLA:
         model.tie_weights()
 
     # We need to recalculate our total training steps as the size of the training dataloader may have changed.