fix

flexflow · Oct 31, 2023 · 617ba79 · 617ba79
1 parent cdc12e6
commit 617ba79
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/inference/utils/download_hf_model.py b/inference/utils/download_hf_model.py
@@ -54,7 +54,7 @@ def main(args):
             llm.download_hf_tokenizer_if_needed()
             llm.download_hf_config()
 
-
+incr_decoding -ll:gpu 1 -ll:cpu 4 -ll:fsize 22000 -ll:zsize 30000 -ll:util 4 -llm-model decapoda-research/llama-7b-hf -prompt ../inference/prompts/chatgpt.json -tensor-parallelism-degree 1 --fusion
 if __name__ == "__main__":
     args = parse_args()
     main(args)