List your company

Question Answering on TriviaQA Wiki (val)

87.6Exact Match (EM)Llama 2
27.69643.24858.874.352Jul 18, 2023Nov 28, 2023Apr 10, 2024Aug 21, 2024Jan 2, 2025May 15, 2025Sep 26, 2025

Evaluation results

MethodLinks
Llama 2Size=70B, Shots=5-shot2023.0787.6
Llama 2Size=70B, Shots=64-shot2023.0787.5
Llama 1Size=65B, Shots=64-shot2023.0786
Llama 1Size=65B, Shots=5-shot2023.0785.9
Llama 2Size=70B, Shots=1-shot2023.0785
Llama 2Size=34B, Shots=64-shot2023.0784.6
Llama 1Size=65B, Shots=1-shot2023.0784.5
Llama 2Size=34B, Shots=5-shot2023.0784.5
Llama 1Size=33B, Shots=5-shot2023.0783.8
Llama 1Size=33B, Shots=64-shot2023.0783.6
Llama 2Size=34B, Shots=1-shot2023.0783.3
Llama 2Size=70B, Shots=0-shot2023.0782.4
Llama 1Size=65B, Shots=0-shot2023.0781.7
Llama 2Size=34B, Shots=0-shot2023.0781
Llama 1Size=33B, Shots=1-shot2023.0780.7
FalconSize=40B, Shots=5-shot2023.0779.9
FalconSize=40B, Shots=64-shot2023.0779.6
Llama 2Size=13B, Shots=5-shot2023.0779.6
Llama 2Size=13B, Shots=64-shot2023.0779.4
Llama 1Size=33B, Shots=0-shot2023.0778.7
FalconSize=40B, Shots=1-shot2023.0778.6
Llama 1Size=13B, Shots=64-shot2023.0777.9
Llama 2Size=13B, Shots=1-shot2023.0777.2
Llama 1Size=13B, Shots=5-shot2023.0777.1
FalconSize=40B, Shots=0-shot2023.0774.6
Llama 1Size=13B, Shots=1-shot2023.0774.4
Llama 2Size=7B, Shots=64-shot2023.0773.7
MPTSize=30B, Shots=64-shot2023.0773.6
MPTSize=30B, Shots=5-shot2023.0773.3
Llama 2Size=13B, Shots=0-shot2023.0773.1
Llama 2Size=7B, Shots=5-shot2023.0772.1
MPTSize=30B, Shots=1-shot2023.0771.3
Llama 1Size=7B, Shots=64-shot2023.0771
Llama 1Size=7B, Shots=5-shot2023.0770.4
Llama 1Size=13B, Shots=0-shot2023.0770.1
Llama 2Size=7B, Shots=1-shot2023.0768.9
MPTSize=30B, Shots=0-shot2023.0768
Llama 1Size=7B, Shots=1-shot2023.0767.4
Llama 2Size=7B, Shots=0-shot2023.0765.8
FalconSize=7B, Shots=5-shot2023.0764.6
Llama 1Size=7B, Shots=0-shot2023.0763.3
MPTSize=7B, Shots=64-shot2023.0761.6
MPTSize=7B, Shots=5-shot2023.0761.2
FalconSize=7B, Shots=64-shot2023.0761.1
MPTSize=7B, Shots=1-shot2023.0759.6
FalconSize=7B, Shots=1-shot2023.0756.8
MPTSize=7B, Shots=0-shot2023.0755.7
FalconSize=7B, Shots=0-shot2023.0752.6
BaselineModel Scale=1B, Evaluation=3-shot2025.0940
InductionModel Scale=1B, Evaluation=3-shot2025.0930
Anti-inductionModel Scale=1B, Evaluation=3-shot2025.0930
BalancedModel Scale=1B, Evaluation=3-shot2025.0930