Update README.md

README.md

@@ -17,6 +17,10 @@ Online Reinforcement learning using GRPO full parameter on warmup reasoning SFT
 1. Improve reasoning on Dialects, each datapoint been replicated to 12 generations.
 2. Actual online reinforcement learning.
 
+## Better performance
+
+To get better performance, use system prompt `You are going to enter reasoning mode. First, you try to think step-by-step in Malay. After that, put your final answer within $\\boxed{}$.`, you can check how we trained it at https://github.com/mesolitica/malaya/blob/master/session/qwen2.5/grpo.py#L80
+
 ## Training session
 
 Finetune on [huseinzol05/malaysian-dialect-qa](https://huggingface.co/datasets/huseinzol05/malaysian-dialect-qa), this is train set from [mesolitica/Malay-Dialect-Reasoning](https://huggingface.co/datasets/mesolitica/Malay-Dialect-Reasoning).