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

    Neuronpedia

    INDEX
    Explanations

    contented sigh

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ند
    0.88
    ।
    0.87
    ي
    0.82
    iam
    0.80
    elt
    0.80
    dır
    0.76
    ^{\
    0.75
    ah
    0.75
    دار
    0.75
    ेड
    0.75
    POSITIVE LOGITS
     sigh
    0.91
     entera
    0.89
     problemática
    0.85
     vecin
    0.82
     admisión
    0.81
     arrondi
    0.80
     occupé
    0.80
    在
    0.79
    要有
    0.78
     iba
    0.78
    Act Density 0.003%

    No Known Activations