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

    Neuronpedia

    INDEX
    Explanations

    interface actions and navigation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     erzählen
    0.86
     verkaufen
    0.74
     konusunda
    0.74
     göst
    0.74
     distribuir
    0.73
     unterstützen
    0.73
     liquido
    0.72
     bestimmen
    0.72
     schützen
    0.72
     الاش
    0.71
    POSITIVE LOGITS
     paste
    1.54
     type
    1.52
    paste
    1.31
    type
    1.28
    Paste
    1.26
     Paste
    1.23
     click
    1.23
     Type
    1.22
    Type
    1.17
     typed
    1.16
    Act Density 0.303%

    No Known Activations