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

    Neuronpedia

    INDEX
    Explanations

    appropriate or inappropriate behavior

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sayesinde
    0.68
    Plus
    0.68
    Gracias
    0.66
     Needed
    0.65
    Available
    0.64
    便于
    0.62
     ईमान
    0.62
     nötig
    0.61
    Forbidden
    0.60
    Required
    0.60
    POSITIVE LOGITS
     practices
    1.15
    erweise
    0.97
     behavior
    0.97
     طریقے
    0.91
     behaviors
    0.91
     Practices
    0.87
     behaviour
    0.86
     behaviours
    0.86
     comportamento
    0.80
    practices
    0.79
    Act Density 0.457%

    No Known Activations