Update README.md

README.md

@@ -1,5 +1,8 @@
-Mistral 7B with step-by-step PPO supervised by [math-shepherd-mistral-7b-prm](https://huggingface.co/peiyi9979/math-shepherd-mistral-7b-prm). We use questions in MetaMATH [1] for PPO training.
-The model is initiated with [mistral-7b-sft](https://huggingface.co/peiyi9979/mistral-7b-sft).
+Step-by-step PPO enhanced [mistral-7b-sft](https://huggingface.co/peiyi9979/mistral-7b-sft).
+
+PRM: [math-shepherd-mistral-7b-prm](https://huggingface.co/peiyi9979/math-shepherd-mistral-7b-prm). 
+
+We use questions in MetaMATH [1] for PPO training.
 
 Pass@1:
 - GSM8K: 84.1