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

    Neuronpedia

    INDEX
    Explanations

    code and technical terms

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Fri
    0.73
    rodní
    0.73
     količ
    0.72
    arthed
    0.71
     সহায়তা
    0.71
    အ
    0.70
     бота
    0.69
     parete
    0.69
    ocking
    0.68
    ນ
    0.68
    POSITIVE LOGITS
    emphasis
    0.76
    чины
    0.68
    hy
    0.63
    kia
    0.60
     Heil
    0.60
     luận
    0.59
    सायन
    0.59
     swords
    0.58
     궁금
    0.58
     toe
    0.58
    Act Density 0.082%

    No Known Activations