Siebe comments on Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?