List your company

Commonsense Reasoning on Winogrande (0-shot and 32-shot evaluation)

73.7Accuracy (0-shot)OPT 175B
47.28454.1426167.858Dec 22, 2022Jun 9, 2023Nov 25, 2023May 12, 2024Oct 28, 2024Apr 15, 2025Oct 2, 2025

Evaluation results

MethodLinks
OPT 175BScale=175B2022.1273.777.6
OPT-IML 175BScale=175B2022.127374.4
DenseSparsity=0%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1072.14—
ALPSSparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1072.14—
L-ADMMSparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1071.98—
ALPSSparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1071.9—
WandaSparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1071.43—
SparseGPTSparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1071.35—
SparseGPTSparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1070.56—
L-ADMMSparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1070.32—
ELSASparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1070—
OPT 30BScale=30B2022.1269.771.7
SAFESparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1069.14—
WandaSparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1068.19—
SAFESparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1067.88—
OPT-IML 30BScale=30B2022.1267.869
ELSASparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1067.8—
MagnitudeSparsity=50%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1065.35—
L-ADMMSparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1065.35—
ALPSSparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1064.56—
SparseGPTSparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1060.93—
ELSASparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1060.38—
OPT 1.3BScale=1.3B2022.1259.556.4
SAFESparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1058.96—
OPT-IML 1.3BScale=1.3B2022.125858.7
MagnitudeSparsity=60%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1057.46—
ALPSSparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1054.93—
ELSASparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1053.75—
SAFESparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1053.43—
L-ADMMSparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1053.28—
MagnitudeSparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1051.78—
WandaSparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1051.54—
WandaSparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1050.83—
SAFESparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1050.51—
ELSASparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1050.43—
WandaSparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1050.04—
L-ADMMSparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1050.04—
SparseGPTSparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1049.49—
SparseGPTSparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1049.33—
MagnitudeSparsity=70%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1049.25—
MagnitudeSparsity=80%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1049.25—
ALPSSparsity=90%, Backbone=Llama-2 13B, Protocol=Zero-shot2025.1048.3—