MIPRO does not return demonstrations

[abedini-arteriaai]

Hi DSPy team, I'm trying to understand how MIPRO functions and have a few questions.

1. Are optimized instructions kept in best_program.trial_logs? In my experiments with GPT3.5, the optimized instructions are saved under extended_signature_instructions whereas best_program.trial_logs only keeps the original instructions for each of the trials.
2. In a similar vein, why keep the optimized instructions and prefix under extended_signature_instructions instead of replacing signature_instructions entirely?

Thank you.