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

    Neuronpedia

    INDEX
    Explanations

    unintended negative outcomes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     suitably
    0.43
     fiducia
    0.42
     truss
    0.42
    culis
    0.41
    Nom
    0.40
     correctement
    0.40
    Gru
    0.40
    Valor
    0.39
    INAL
    0.38
    Deux
    0.38
    POSITIVE LOGITS
     unwanted
    0.96
    してしまう
    0.89
     unintended
    0.86
     undesired
    0.86
     uncontroll
    0.85
     unintentionally
    0.85
     uncontrollable
    0.84
     undesirable
    0.83
    ってしまう
    0.82
     inadvertently
    0.77
    Act Density 0.258%

    No Known Activations