Skip to content

Benchmark

HF Hallucinations TruthfulQA MC2, Accuracy

Factuality benchmarks measure how often models get facts right. Higher scores are better.

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

0.446
Mean
0.453
Median
0.007
Std dev
0.439
10th percentile
0.453
90th percentile
0.439
Min

Leaderboard

#ModelScore
1Llama-2-7b-chat-hfclustermap_all/TruthfulQA MC2, Accuracy· HF Hallucinations aggregate matrix0.453
2Llama-2-13b-chat-hfclustermap_all/TruthfulQA MC2, Accuracy· HF Hallucinations aggregate matrix0.439

Data credits: Hallucinations Leaderboard, Apache-2.0. Publisher-selected archived score matrix; native normalized fractions. (licence)