The gs_design_ahr function calculates the ~HR at the bound using the formula exp(-z / sqrt(info0)), while gs_power_ahr derives it using exp(-z / sqrt(info)). I will check the formula and fix the inconsistency.

References:

• https://github.com/Merck/gsDesign2/blob/main/R/gs_design_ahr.R#L350
• https://github.com/Merck/gsDesign2/blob/main/R/gs_power_ahr.R#L250