Skip to content

Benchmark

HF Hallucinations SQuADv2, EM

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

0.126
Mean
0.133
Median
0.006
Std dev
0.12
10th percentile
0.133
90th percentile
0.12
Min

Leaderboard

#ModelScore
1Llama-2-13b-chat-hfclustermap_all/SQuADv2, EM· HF Hallucinations aggregate matrix0.133
2Llama-2-7b-chat-hfclustermap_all/SQuADv2, EM· HF Hallucinations aggregate matrix0.12

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