Where does action clipping happen?

[51616]

Hey, thank you for such a great addition to multi-agent cooperative environment.

I am playing with the environment and notice that the environment's action space is bounded within [-1,1]. But if the action is outside the range, the environment doesn't throw any error. Instead, it works normally. So I wonder where does the clipping happen?

Another question regarding the FACMAC paper. Did you use any wrapper such as observation/reward normalization or action clipping/rescaling? 'Cause in the original single-agent mode, the implementation usually use normalization and clipping wrapper for Mujoco tasks.