fix bug of release output in pp (#54624)

python/paddle/distributed/fleet/meta_parallel/pipeline_parallel.py

@@ -671,11 +671,19 @@ class PipelineParallel(MetaParallelBase):
             self.lr_scheduler.step()
 
     def _release_output(self, output):
+        def can_free(t):
+            return (
+                t is not None
+                and isinstance(t, paddle.Tensor)
+                and t.inplace_version == 0
+            )
+
         if isinstance(output, (tuple, list)):
             for t in output:
-                if t is not None and isinstance(t, paddle.Tensor):
+                if can_free(t):
                     t._clear_dataptr()
-        elif output is not None and isinstance(output, paddle.Tensor):
+
+        elif can_free(output):
             output._clear_dataptr()