Meaning of training process results in tensorboard

Hi all,

as someone who is just trying to introduce RL to myself, it would be great if I could somehow find the explaination for the plots in tensorboard:

Is it documented somewhere and I am just missing it?

Hi, pushing this back to the top of the list, because I have the same question. Can anybody help?

I found out that I can add to the tensorboard by adding key-value pairs to the “extras” dict. But I did not find out where the entries for losses, performance, info and rewards are defined.