Which hyper-parameters in the standard VERL configuration do not take effect?

[dalek-who]

This is the full VERL config, https://verl.readthedocs.io/en/latest/examples/config.html
I noticed that some parameters seem not work when passed to agl.VERL. For example:

• actor_rollout_ref.rollout.val_kwargs
• data.truncation
• actor_rollout_ref.rollout.top_p (I'm not sure if it takes effect, as it is not explicitly set like actor_rollout_ref.rollout.temperature)

agent-lightning/agentlightning/verl/trainer.py

Lines 463 to 469 in 5f3093d

	self.agent_mode_daemon = self.daemon_cls(
	self.config.agentlightning.port,
	self.config.actor_rollout_ref.rollout.n,
	train_information={
	"model": model,
	"temperature": self.config.actor_rollout_ref.rollout.temperature,
	},

Could you provide a blacklist/whitelist to clarify which configs are effective or not?

[ultmaster]

That's a good suggestion. Maybe we can go through all the parameters and make it more clear!