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

    Neuronpedia

    INDEX
    Explanations

    gentle, delicate, sensitive, careful actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    som
    1.92
    க்கொண்டு
    1.86
     entier
    1.81
    в
    1.79
     enige
    1.71
    きた
    1.70
    রু
    1.67
    ma
    1.67
     daha
    1.65
    ад
    1.64
    POSITIVE LOGITS
     sekali
    1.70
    ﺔ
    1.69
    पूर्वक
    1.67
    आइए
    1.66
    ":"",
    1.66
     fring
    1.65
    TE
    1.64
     linewidth
    1.62
    1.62
     refunds
    1.58
    Act Density 0.539%

    No Known Activations