diff --git a/server/text_generation_server/models/__init__.py b/server/text_generation_server/models/__init__.py index 3e181321..85f0d15c 100644 --- a/server/text_generation_server/models/__init__.py +++ b/server/text_generation_server/models/__init__.py @@ -147,7 +147,7 @@ def get_model( ) elif model_type == "gpt_neox": - if FLASH_ATTENTION or shard: + if FLASH_ATTENTION: return FlashNeoXSharded( model_id, revision,