Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    hesitation and filler words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ißler
    0.39
    петров
    0.38
     gór
    0.37
    üedad
    0.37
    namespaces
    0.37
    जीटी
    0.37
     Kerk
    0.37
    чай
    0.36
    കേ
    0.36
    OLO
    0.36
    POSITIVE LOGITS
     yeah
    0.71
     huh
    0.68
    …
    0.64
    hh
    0.62
    ...
    0.59
     oh
    0.58
    hmm
    0.58
    ..
    0.56
     hm
    0.55
     hmm
    0.54
    Act Density 0.008%

    No Known Activations