Skip to content

Revert "[reward] fix: fix reward computation in _validate when use_r… #1

Revert "[reward] fix: fix reward computation in _validate when use_r…

Revert "[reward] fix: fix reward computation in _validate when use_r… #1

Triggered via push January 29, 2026 01:46
Status Success
Total duration 7s
Artifacts
setup
setup
e2e_ppo_trainer_fsdp_sglang
0s
e2e_ppo_trainer_fsdp_sglang
e2e_ppo_trainer_fsdp-qwen2_5vl-3b
e2e_ppo_trainer_fsdp-qwen2_5vl-3b
cleanup
4s
cleanup
Fit to window
Zoom out
Zoom in