Have you considered skipping that last step? (the asking chatgpt at all part)
I don't know if I'm just a freaky outlier in some way, or if my natural animosity towards the technology is blinding me to its benefits, but I really don't get people using LLMs for analysis....
Fair enough- llms are not known for their accuracy. But in addition to the other data, I think it’s an interesting point that despite their fine tuning to be as neutral/tame-green-woke as possible, they couldn’t help but begrudgingly admit it (when I asked a few a while...