I would like to ask for some advice. I used the nuPlan dataset to perform SFT and RFT training. However, during the final testing stage, the generated token IDs never exceed 151665, so the action tokens cannot be parsed correctly. What could be the possible cause of this issue?
I would like to ask for some advice. I used the nuPlan dataset to perform SFT and RFT training. However, during the final testing stage, the generated token IDs never exceed 151665, so the action tokens cannot be parsed correctly. What could be the possible cause of this issue?