List your company

Instruction Following on AlpacaEval (805 prompts)

7.22OLMo-2 Reward ScoreAAD
1.28162.82334.3655.9067Sep 30, 2025

Evaluation results

MethodLinks
AADBackbone=OLMo-2-13B2025.097.2298—
EFTBackbone=OLMo-2-13B2025.096.979655.8
Bo2Backbone=OLMo-2-13B2025.096.949665.2
Greedy DPOBackbone=OLMo-2-13B2025.096.879575.2
Greedy SFTBackbone=OLMo-2-13B2025.095.02—93.9
AADBackbone=OLMo-2-7B2025.093.8498—
EFTBackbone=OLMo-2-7B2025.093.59658.1
Bo2Backbone=OLMo-2-7B2025.093.449764.6
Greedy DPOBackbone=OLMo-2-7B2025.093.279471.5
Greedy SFTBackbone=OLMo-2-7B2025.091.51—91.6