训练过程中 Tensorboard Log 中的 Rollout Entropy 和 Entropy 差距极大 #1574

Open

opened

on Feb 11, 2026

如题
训练过程中 rollout/entropy 和 train/entropy 之间差好几倍，是哪里计算出了问题么？

Metadata

Assignees

No one assigned

Labels

No labels

No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests