CoolFace
Modelpublic

payelb/PKUSafeRLHF_reward-model-deberta-v3-base_1k_fixed_adaboost_margin_noaug

sourceHugging Faceupdated 6mo agoView on Hugging Face
0likes8downloads
4 commits on main
b1f381e6mo ago

final: adaboost-style fixed1k no-aug margin-weighted RM (PKUSafeRLHF, rounds=5, tau=5, lr=2e-05, gas=16)

payelb
537a59a6mo ago

final: adaboost-style fixed1k no-aug margin-weighted RM (PKUSafeRLHF, rounds=5, tau=1.0, lr=2e-05, gas=16)

payelb
6bac6586mo ago

final: adaboost-style fixed1k no-aug margin-weighted RM (PKUSafeRLHF, rounds=5, tau=1.0, lr=2e-05, gas=16)

payelb
02e296c6mo ago

initial commit

payelb