List your company

Honesty alignment on HHH alignment

86.1Win Rate (WR)AMBS
23.80439.97756.1572.323Sep 26, 2025

Evaluation results

MethodLinks
AMBSBackbone=DeepSeek-7B, Throughput (Tk)=230, Memory Usage (Me)=19, Compute Cost (GP)=42025.0986.130.867.540.9
AMBSBackbone=Mistral-7B, Throughput (Tk)=218, Memory Usage (Me)=20, Compute Cost (GP)=52025.0951.830.146.122.6
AMBSBackbone=Gemma-7B, Throughput (Tk)=229, Memory Usage (Me)=20, Compute Cost (GP)=52025.0950.422.132.220.1
SteeringBackbone=Mistral-7B, Throughput (Tk)=215, Memory Usage (Me)=22, Compute Cost (GP)=62025.0936.43335.913.1
AlignerBackbone=Mistral-7B, Throughput (Tk)=182, Memory Usage (Me)=26, Compute Cost (GP)=82025.0935.136.135.311.4
AMBSBackbone=LLaMA-2-7B, Throughput (Tk)=210, Memory Usage (Me)=21, Compute Cost (GP)=62025.0934.93037.914.3
DExpertsBackbone=Mistral-7B, Throughput (Tk)=179, Memory Usage (Me)=27, Compute Cost (GP)=92025.0934.63434.911.8
PPLMBackbone=Mistral-7B, Throughput (Tk)=155, Memory Usage (Me)=29, Compute Cost (GP)=102025.0932.834.933.210.4
SteeringBackbone=Gemma-7B, Throughput (Tk)=221, Memory Usage (Me)=21, Compute Cost (GP)=62025.0932.325.727.811.5
AlignerBackbone=Gemma-7B, Throughput (Tk)=188, Memory Usage (Me)=25, Compute Cost (GP)=72025.0931.528.727.910.2
DExpertsBackbone=Gemma-7B, Throughput (Tk)=185, Memory Usage (Me)=26, Compute Cost (GP)=82025.0930.926.827.210.4
PPLMBackbone=Gemma-7B, Throughput (Tk)=161, Memory Usage (Me)=28, Compute Cost (GP)=92025.0929.727.426.19.5
SteeringBackbone=LLaMA-2-7B, Throughput (Tk)=209, Memory Usage (Me)=23, Compute Cost (GP)=72025.0929.431.732.610.1
SteeringBackbone=DeepSeek-7B, Throughput (Tk)=227, Memory Usage (Me)=20, Compute Cost (GP)=62025.0928.932.548.214.9
AlignerBackbone=LLaMA-2-7B, Throughput (Tk)=176, Memory Usage (Me)=27, Compute Cost (GP)=92025.0928.634.93.9-0.8
AlignerBackbone=DeepSeek-7B, Throughput (Tk)=194, Memory Usage (Me)=24, Compute Cost (GP)=72025.092835.648.813.7
DExpertsBackbone=LLaMA-2-7B, Throughput (Tk)=173, Memory Usage (Me)=28, Compute Cost (GP)=102025.0927.932.631.79
DExpertsBackbone=DeepSeek-7B, Throughput (Tk)=191, Memory Usage (Me)=25, Compute Cost (GP)=82025.0927.433.247.513.9
PPLMBackbone=LLaMA-2-7B, Throughput (Tk)=149, Memory Usage (Me)=30, Compute Cost (GP)=112025.0926.533.530.77.9
PPLMBackbone=DeepSeek-7B, Throughput (Tk)=167, Memory Usage (Me)=27, Compute Cost (GP)=92025.0926.23446.713