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

    Neuronpedia

    INDEX
    Explanations

    por

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     myſelf
    -0.75
     Efq
    -0.72
     greateſt
    -0.72
     Anſ
    -0.71
     themſelves
    -0.70
     whoſe
    -0.70
     Houſe
    -0.69
     itſelf
    -0.69
     Abbé
    -0.68
     becauſe
    -0.67
    POSITIVE LOGITS
    routeProvider
    0.52
    iar
    0.50
     lift
    0.45
    ikon
    0.45
    nelson
    0.45
    a
    0.45
    elect
    0.45
    i
    0.45
    ion
    0.43
    ik
    0.43
    Act Density 0.065%

    No Known Activations