How closely do you want to match the mean of 5? Here is one way to get close but not exactly 5.
set.seed(1)
DAT <- rnorm(98, mean = 5, sd = 5/3)
range(DAT)
#> [1] 1.308834 9.002696
mean(DAT)
#> [1] 5.21406
DAT <- c(DAT, 0, 10)
range(DAT)
#> [1] 0 10
mean(DAT)
#> [1] 5.209779
qqnorm(DAT)

shapiro.test(DAT)
#>
#> Shapiro-Wilk normality test
#>
#> data: DAT
#> W = 0.98965, p-value = 0.6364
Created on 2021-02-24 by the reprex package (v0.3.0)