List your company

Commonsense Reasoning on Capability Evaluation Suite Commonsense Pool

97.03S_Com Scoremix_balanced
96.062896.313996.56596.8161Jun 2, 2026

Evaluation results

MethodLinks
mix_balancedSparsity=40%, Pruning Method=Wanda, Base Model=LLaMA-3.1-8B2026.0697.03
C4 (random)Sparsity=40%, Pruning Method=Wanda, Base Model=LLaMA-3.1-8B2026.0696.39
SGS-styleSparsity=40%, Pruning Method=Wanda, Base Model=LLaMA-3.1-8B2026.0696.2
Self-Gen (from C4)Sparsity=40%, Pruning Method=Wanda, Base Model=LLaMA-3.1-8B2026.0696.1