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

    Neuronpedia

    INDEX
    Explanations

    difficulties and downsides

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gpointer
    0.43
     isValid
    0.40
     erkennen
    0.39
    න්ත
    0.38
     manually
    0.38
    NORMAL
    0.38
    建立了
    0.37
     enzyme
    0.37
     বলিয়
    0.37
    isValid
    0.37
    POSITIVE LOGITS
     downsides
    0.65
     restrictions
    0.62
     terlalu
    0.59
     싫
    0.59
     overkill
    0.59
     restricciones
    0.58
     inconvén
    0.57
     불편
    0.56
     hassles
    0.56
     pricey
    0.54
    Act Density 0.133%

    No Known Activations