top | item 47008792 (no title) kfajdsl | 16 days 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. discuss order hn newest energy123|16 days 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"
energy123|16 days 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"
energy123|16 days ago