Prompt Engineered GPT-4 Beats Gemini on all of Google's text benchmarks

[sp332]

"We note that Medprompt+ relies on accessing confidence scores (logprobs) from GPT-4. These are not publicly available via the current API but will be enabled for all in the near future."

Had that been announced yet?

[alecco]

It would be interesting to see how this can be used for open models.

[byyoung3]

engineered? Over overfit?