Fix for new versions

custom_generate/generate.py

@@ -179,7 +179,9 @@ def _dola_decoding(
         # keep track of which sequences are already finished
         batch_size, cur_length = input_ids.shape[:2]
         unfinished_sequences = torch.ones(batch_size, dtype=torch.long, device=input_ids.device)
-        model_kwargs = model._get_initial_cache_position(cur_length, input_ids.device, model_kwargs)
+        # Does not exist anymore in recent versions!
+        if hasattr(model, "_get_initial_cache_position"):
+            model_kwargs = model._get_initial_cache_position(cur_length, input_ids.device, model_kwargs)
 
         this_peer_finished = False