Common Sense Reasoning On Arc Easy

Accuracy

Results

Performance results of various models on this benchmark

		Paper Title
ST-MoE-32B 269B (fine-tuned)	95.2	ST-MoE: Designing Stable and Transferable Sparse Expert Models
LLaMA 3 8B+MoSLoRA (fine-tuned)	90.5	Mixture-of-Subspaces in Low-Rank Adaptation
PaLM 2-L (1-shot)	89.7	PaLM 2 Technical Report
PaLM 2-M (1-shot)	88.0	PaLM 2 Technical Report
LLaMA-3 8B + MixLoRA	86.5	MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of Experts
Camelidae-8×34B	86.2	Parameter-Efficient Sparsity Crafting from Dense to Mixture-of-Experts for Instruction Tuning on General Tasks
PaLM 2-S (1-shot)	85.6	PaLM 2 Technical Report
LLaMA 65B + CFG (0-shot)	84.2	Stay on topic with Classifier-Free Guidance
GAL 120B (0-shot)	83.8	Galactica: A Large Language Model for Science
LLaMA-2 13B + MixLoRA	83.5	MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of Experts
LLaMA 30B + CFG (0-shot)	83.2	Stay on topic with Classifier-Free Guidance
Mixtral 8x7B (0-shot)	83.1	Mixtral of Experts
FLAN 137B (few-shot, k=14)	80.7	Finetuned Language Models Are Zero-Shot Learners
Mistral 7B (0-shot)	80.5	Mixtral of Experts
LLaMA 33B (0-shot)	80.0	LLaMA: Open and Efficient Foundation Language Models
Mistral 7B (0-shot)	80.0	Mistral 7B
FLAN 137B (0-shot)	79.6	Finetuned Language Models Are Zero-Shot Learners
LLaMA 13B + CFG (0-shot)	79.1	Stay on topic with Classifier-Free Guidance
LLaMA 65B (0-shot)	78.9	LLaMA: Open and Efficient Foundation Language Models
LLaMA-2 7B + MixLoRA	77.7	MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of Experts

0 of 47 row(s) selected.