Running minimize with Lammps will repeatedly calculate the model deviation

[jpqiu]

When I run the minimize operation in Lammps, the log.lammps file outputs normally, but I've noticed that the model_devi.out file exhibits repetitive calculations. I'm trying to understand the cause of this issue. For instance, while log.lammps indicates that 567 steps have been executed, the model_devi.out file contains 1129 lines.Furthermore, the model deviation calculated for the same number of steps is not consistent.
I am using version 2.2.7. Below are my input settings along with the log.lammps and model_devi.out files.

input_output.zip

[jpqiu]

It's possible that the CG algorithm also outputs the model deviation of internal calculations, which is not conducive to my sampling. Is there any way to correct this?

[njzjz]

We rejected the idea to distinguish the internal calculations in #1311 due to side effects.

[jpqiu]

Thank you for your reply