Look what they need to mimic a fraction of [the power of having the logit probab... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

matjet 4 months ago | parent | context | favorite | on: Gemini 3 Deep Think

Look what they need to mimic a fraction of [the power of having the logit probabilities exposed so you can actually see where the model is uncertain]

kfajdsl 4 months ago [–]

All the LLM logprob outputs I've seen aren't very well calibrated, at least for transcription tasks - I'm guessing it's similar for OCR type tasks.

energy123 4 months ago | [–]

"I already decided in my private reasoning trace to resolve this ambiguity by emitting the string '27' instead of '22' right here, thus '27' has 100% probability"

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact