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

    Neuronpedia

    INDEX
    Explanations

    instincts and masked behavior

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    рий
    0.54
    ᙱ
    0.47
     Е
    0.46
     нажа
    0.44
    ULD
    0.44
    getattr
    0.43
     પ્રિય
    0.43
     있던
    0.43
     ಎಸ್
    0.43
    Clipboard
    0.42
    POSITIVE LOGITS
    H
    0.48
     ALUMIN
    0.45
    ”)
    0.44
    jeu
    0.44
     philanth
    0.44
    真っ
    0.44
    沒
    0.44
     té
    0.42
    </b>
    0.42
     niez
    0.42
    Act Density 0.001%

    No Known Activations