List your company
ResearchFormerly WizWand

General Knowledge Evaluation on C-Eval (val)

34.32AccuracyLLaMA2-13B
22.422425.511228.631.6888Jul 23, 2024Nov 6, 2024Feb 20, 2025Jun 7, 2025Sep 21, 2025Jan 5, 2026Apr 22, 2026
Updated 6mo ago

Evaluation results

MethodLinks
LLaMA2-13BModel Size=13B, Shots=0-shot2024.0734.32
TinyLLaMA-1.1B + DDKModel Size=1.1B, Shots=0-shot2024.0727.86
TinyLLaMA-1.1B + TEDModel Size=1.1B, Shots=0-shot2024.0727.49
TinyLLaMA-1.1B + KDModel Size=1.1B, Shots=0-shot2024.0727.12
TinyLLaMA-1.1B + CPTModel Size=1.1B, Shots=0-shot2024.0726.79
TinyLLaMA-1.1B + MiniLLMModel Size=1.1B, Shots=0-shot2024.0726.74
TinyLLaMA-1.1B + CPT & DoReMiModel Size=1.1B, Shots=0-shot2024.0726.37
MLFiltering Strategy=ML2026.0426.15
MKC-eFiltering Strategy=MKC-e2026.0425.56
HQFiltering Strategy=HQ2026.0424.89
ML (Q3)Filtering Strategy=ML (Q3)2026.0424.67
HQ seedFiltering Strategy=HQ seed2026.0424.44
TinyLLaMA-1.1BModel Size=1.1B, Shots=0-shot2024.0723.92
ML (15%)Filtering Strategy=ML (15%)2026.0423.7
No filteringFiltering Strategy=No filtering2026.0422.88