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

    Neuronpedia

    INDEX
    Explanations

    bolded text followed by colon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ه
    2.05
    o
    1.92
    ли
    1.76
    ம்
    1.69
    л
    1.67
    ة
    1.44
    ть
    1.43
    en
    1.42
    े
    1.40
    ת
    1.40
    POSITIVE LOGITS
    namely
    1.25
    кість
    1.22
    estructive
    1.18
    㗆
    1.15
    mainWindow
    1.12
     Maladies
    1.09
    儡
    1.06
     Aa
    1.06
    <td>
    1.05
    derived
    1.05
    Act Density 0.135%

    No Known Activations