payelb/PKUSafeRLHF_reward-model-deberta-v3-base_1k_fixed_MARS_semantic_refined
010
final: PKU MARS semantic-refined fixed1k T=5, alpha=0.5, B=5000, lr=2e-05, gas=16
final: PKU MARS semantic-refined fixed1k T=5, alpha=0.5, B=5000, lr=2e-05, gas=16
final: PKU MARS semantic-refined fixed1k T=5, alpha=0.5, B=5000, lr=2e-05, gas=16
initial commit
