[fix] type error of quantile

[gitabtion]

What does this PR do?

Entropies is a bfloat16 tensor when training with bf16, but quantile() input tensor must be either float or double dtype.

trl/trl/trainer/grpo_trainer.py

Line 1395 in 6a6d434

entropy_threshold = torch.quantile(entropies.flatten(), self.token_entropy_percentile_threshold)

Fixes #3666

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a GitHub issue? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

[kashif]

thanks @gitabtion are you using torch<=2.6?

[gitabtion]

@kashif Yes, "torch==2.6.0+cu124".