Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Question Answering on TruthfulQA (Truthful/Inf Metrics)

88.23Truthful*Inf Score

Llama2-7B

16.230834.922953.61572.3071Feb 27, 2023Apr 28, 2023Jun 28, 2023Aug 28, 2023Oct 28, 2023Dec 28, 2023Feb 27, 2024
Updated 4d ago

Evaluation Results

MethodLinks
2023.11
88.2395.192.78-
2023.11
81.9293.8887.27-
2023.11
76.9278.9597.430.7897
2023.11
71.9674.4296.70.8796
2023.11
67.8769.6597.450.7396
2024.02
65.4572.9589.72-
2023.11
64.8174.0587.520.7417
2023.11
62.5470.3888.860.4855
2023.11
61.1469.2888.250.8848
2023.11
59.6168.387.270.4647
2023.11
58.864.1491.680.9555
2024.02
54.5667.4480.91-
2023.02
5357--
2023.11
50.3955.8395.10.8659
2023.07
50.18---
2023.07
48.71---
2023.11
48.257.1684.330.8416
2023.02
4852--
2024.02
44.455.380.29-
2023.07
44.19---
2023.07
43.45---
2023.11
42.5945.992.780.4931
2024.02
42.2364.3865.59-
2023.07
41.86---
2023.07
41.74---
2023.11
41.4344.0693.760.5051
2024.02
41.3842.198.3-
2023.02
4147--
2023.07
40.39---
2024.02
36.147.176.65-
2023.07
35.25---
2024.02
33.434.796.25-
2023.07
33.29---
2024.02
32.4441.7477.72-
2024.02
31.936.9686.29-
2023.07
29.13---
2023.02
2933--
2023.07
27.42---
2023.07
25.95---
2023.02
2528--
2023.02
1931--
2023.02
1922--