List your company

Instruction Following on AlpacaFarm Eval (test)

76.13Win RateBest-of-N
48.871655.948363.02570.1017Oct 26, 2024

Evaluation results

MethodLinks
Best-of-NN=3840, Base Model=Mistral-7B2024.1076.1377.21
Best-of-NN=1920, Base Model=Mistral-7B2024.1075.276.27
SPECULATIVE REJECTIONalpha=0.5, Base Model=Llama-3-8B2024.1073.677.91
SPECULATIVE REJECTIONalpha=0.5, Base Model=Mistral-7B2024.1069.4273.31
Best-of-NN=960, Base Model=Mistral-7B2024.1067.568.07
SPECULATIVE REJECTIONalpha=0.5, Base Model=Average2024.1066.1770.01
Best-of-NN=3840, Base Model=Average2024.1062.8963.04
Best-of-NN=1920, Base Model=Average2024.1062.5463.48
Best-of-NN=480, Base Model=Mistral-7B2024.1061.2861.84
Best-of-NN=240, Base Model=Mistral-7B2024.1060.6960.07
Best-of-NN=1920, Base Model=Llama-3-8B2024.1060.5761.05
Best-of-NN=960, Base Model=Llama-3-8B2024.1059.260.26
Best-of-NN=3840, Base Model=Llama-3-8B2024.1059.1957.91
Best-of-NN=960, Base Model=Average2024.1059.0359.99
Best-of-NN=480, Base Model=Llama-3-8B2024.1058.959.93
Best-of-NN=480, Base Model=Average2024.1056.8958.29
SPECULATIVE REJECTIONalpha=0.5, Base Model=Llama-3-8B-Instruct2024.1055.558.8
Best-of-NN=240, Base Model=Average2024.1053.6954.41
Best-of-NN=3840, Base Model=Llama-3-8B-Instruct2024.1053.3654.01
Best-of-NN=1920, Base Model=Llama-3-8B-Instruct2024.1051.8653.13
Best-of-NN=480, Base Model=Llama-3-8B-Instruct2024.1050.4953.11
Best-of-NN=240, Base Model=Llama-3-8B2024.1050.4550.27
Best-of-NN=960, Base Model=Llama-3-8B-Instruct2024.1050.3951.64
Best-of-NN=120, Base Model=Mistral-7B2024.105050
Best-of-NN=120, Base Model=Llama-3-8B2024.105050
Best-of-NN=120, Base Model=Llama-3-8B-Instruct2024.105050
Best-of-NN=120, Base Model=Average2024.105050
Best-of-NN=240, Base Model=Llama-3-8B-Instruct2024.1049.9252.89