List your company
ResearchFormerly WizWand

Commonsense Reasoning on HellaSwag 10-shot

83.3ScoreFP16
59.213665.466871.7277.9732Nov 14, 2025Dec 11, 2025Jan 7, 2026Feb 4, 2026Mar 3, 2026Mar 30, 2026Apr 27, 2026
Updated 6mo ago

Evaluation results

MethodLinks
FP16Model=Mistral-7B, Quantization=FP162026.0483.3
COVERCALModel=Mistral-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0482.8
FP16Model=LLaMA-3-8B, Quantization=FP162026.0482.1
Max-ActVarModel=Mistral-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0482.1
StratifiedModel=Mistral-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0482
RandomModel=Mistral-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0481.9
Max-PPLModel=Mistral-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0481.7
COVERCALModel=LLaMA-3-8B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0481.5
Max-ActVarModel=LLaMA-3-8B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0480.8
StratifiedModel=LLaMA-3-8B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0480.7
RandomModel=LLaMA-3-8B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0480.5
Max-PPLModel=LLaMA-3-8B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0480.3
FP16Model=LLaMA-2-7B, Quantization=FP162026.0478.6
COVERCALModel=LLaMA-2-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0478
Max-ActVarModel=LLaMA-2-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0477.4
StratifiedModel=LLaMA-2-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0477.3
RandomModel=LLaMA-2-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0477.1
Max-PPLModel=LLaMA-2-7B, Quantization=AWQ INT4, K (Calibration Samples)=1282026.0477
UM-190kBase Model=OLMo-2-7B-SFT, DPO Training Dataset=UM-190k2025.1162.86
UM-187kBase Model=OLMo-2-7B-SFT, DPO Training Dataset=UM-187k2025.1162.54
UM-170kBase Model=OLMo-2-7B-SFT, DPO Training Dataset=UM-170k2025.1162.17
TuluDPOBase Model=OLMo-2-7B-SFT, DPO Training Dataset=TuluDPO2025.1161.97
ORPOBase Model=OLMo-2-7B-SFT, DPO Training Dataset=ORPO2025.1161.3
HelpSteerBase Model=OLMo-2-7B-SFT, DPO Training Dataset=HelpSteer2025.1160.55
UltraFBBase Model=OLMo-2-7B-SFT, DPO Training Dataset=UltraFB2025.1160.24
CodePrefBase Model=OLMo-2-7B-SFT, DPO Training Dataset=CodePref2025.1160.24
SFTBase Model=OLMo-2-7B-SFT, DPO Training Dataset=SFT2025.1160.14