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

    Neuronpedia

    INDEX
    Explanations

    two options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    煩
    0.92
    堌
    0.91
     fatorial
    0.90
    ية
    0.88
     splic
    0.88
     excret
    0.88
     TInner
    0.88
     elastomer
    0.88
     adsorbent
    0.86
    ন্তন
    0.86
    POSITIVE LOGITS
    1.25
    א
    0.97
    но
    0.95
    begin
    0.86
    dans
    0.86
     was
    0.85
     треть
    0.85
    lif
    0.83
    label
    0.83
    title
    0.82
    Act Density 0.001%

    No Known Activations