From be11ea0fb2bbf7e85859009f92051d29764dfeb2 Mon Sep 17 00:00:00 2001 From: Bo Liu Date: Thu, 19 Dec 2024 13:10:35 +0000 Subject: [PATCH] fix: update is_ppo flag --- nlrl/policy/llm_policy.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/nlrl/policy/llm_policy.py b/nlrl/policy/llm_policy.py index 281c5f3..ff7ebc7 100644 --- a/nlrl/policy/llm_policy.py +++ b/nlrl/policy/llm_policy.py @@ -26,7 +26,7 @@ def __init__( self.tp_size = model_tp_size self.sample_config = sample_config self.is_gpt4 = "gpt-4" in model_path.lower() - self.is_ppo = "ppo" in model_path.lower() + self.is_ppo = "-ppo" in model_path.lower() print(f"model_path: {model_path}, is_gpt4: {self.is_gpt4}, is_ppo: {self.is_ppo}") if self.is_gpt4: