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

    Neuronpedia

    INDEX
    Explanations

    fairness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     destructor
    0.43
    椃
    0.38
    Password
    0.36
    হাম্মদ
    0.35
    stoke
    0.35
     Moc
    0.34
    idencia
    0.34
    啶
    0.34
    uan
    0.34
     hola
    0.34
    POSITIVE LOGITS
     fair
    4.44
    Fair
    4.31
     Fair
    4.25
    fair
    4.13
     fairness
    3.69
     FAIR
    3.66
    公平
    3.59
     Fairness
    3.45
     fairer
    3.30
    フェア
    3.23
    Act Density 0.053%

    No Known Activations