Skip to content

Replace kl_penalty_reference_step with kl_penalty_step_lag#625

Open
angkywilliam wants to merge 13 commits into
feat/pipeline-klfrom
feat/pipeline-kl-step-lag
Open

Replace kl_penalty_reference_step with kl_penalty_step_lag#625
angkywilliam wants to merge 13 commits into
feat/pipeline-klfrom
feat/pipeline-kl-step-lag

feat: Add TinkerNativeBackend yes-no-maybe KL advantage script

ae9e463
Select commit
Loading
Failed to load commit list.