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

    Neuronpedia

    INDEX
    Explanations

    gate and fence actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Platform
    0.59
    Platform
    0.55
     платфор
    0.54
     प्लेटफार्म
    0.52
     platform
    0.48
     플랫폼
    0.47
     plateforme
    0.47
    平台的
    0.47
     प्रोग्राम
    0.46
     платформа
    0.45
    POSITIVE LOGITS
     gate
    1.09
     fence
    1.09
     gates
    1.05
     fences
    0.93
    fence
    0.93
     fencing
    0.92
     Fence
    0.91
     Gate
    0.90
    Fence
    0.88
     Gates
    0.87
    Act Density 0.014%

    No Known Activations