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

    Neuronpedia

    INDEX
    Explanations

    modify or transform

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    als
    1.01
    al
    0.96
    on
    0.92
    foot
    0.90
    ão
    0.87
    ah
    0.87
    ům
    0.84
    um
    0.84
     skrz
    0.83
    ons
    0.83
    POSITIVE LOGITS
     Expand
    0.98
     Mod
    0.96
     De
    0.93
     Redirect
    0.86
     S
    0.86
     Expanding
    0.85
     Сим
    0.83
     Ло
    0.83
     Transforming
    0.82
     Customize
    0.81
    Act Density 0.008%

    No Known Activations