6 ms·
Prompt Engineered GPT-4 Beats Gemini on all of Google's text benchmarks
- byyoung3 3y agoengineered? Over overfit?
- sp332 3y ago"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 3y agoIt would be interesting to see how this can be used for open models.