Skip to content

fix(ppo): exclude no-eos rows from reward normalization#1351

Open
haoyang9804 wants to merge 2 commits into
areal-project:mainfrom
haoyang9804:fix/reward-norm-no-eos-mask
Open

fix(ppo): exclude no-eos rows from reward normalization#1351
haoyang9804 wants to merge 2 commits into
areal-project:mainfrom
haoyang9804:fix/reward-norm-no-eos-mask

Commits

Commits on May 19, 2026