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

    Neuronpedia

    INDEX
    Explanations

    convert percentage to decimal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ={[
    0.66
    ce
    0.64
    MenuItem
    0.62
     OUT
    0.61
    Out
    0.61
     outs
    0.60
    च्छ
    0.60
    ücken
    0.60
     Out
    0.59
    Bj
    0.59
    POSITIVE LOGITS
     догово
    0.73
    provided
    0.71
    łaszcza
    0.71
    躾
    0.70
     જોડ
    0.68
     entusiasmo
    0.67
    ஒரு
    0.66
     Roofing
    0.66
    거든요
    0.65
    asional
    0.65
    Act Density 0.001%

    No Known Activations