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

    Neuronpedia

    INDEX
    Explanations

    fitting descriptions or criteria

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Separate
    0.69
    iert
    0.66
     اث
    0.66
    ieme
    0.65
     الشيء
    0.64
    ҥ
    0.64
     maaf
    0.63
     الصين
    0.63
     thing
    0.63
    把
    0.63
    POSITIVE LOGITS
     snugly
    1.08
     snug
    0.88
     larga
    0.87
    rockets
    0.82
     comfortably
    0.82
    ters
    0.80
    memory
    0.80
    teras
    0.79
    条件的
    0.77
     справо
    0.77
    Act Density 0.118%

    No Known Activations