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

    Neuronpedia

    INDEX
    Explanations

    1 million tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    dod
    0.38
    ४
    0.37
    Fram
    0.35
    Software
    0.34
    ೪
    0.34
     Fram
    0.34
    עם
    0.33
     shares
    0.33
     Pumps
    0.33
    do
    0.33
    POSITIVE LOGITS
     toa
    0.39
    urous
    0.37
     porcel
    0.37
     contenant
    0.37
     escalar
    0.37
    上限
    0.36
    ダウン
    0.36
     વિશ
    0.36
     extender
    0.36
    asma
    0.35
    Act Density 0.008%

    No Known Activations