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

    Neuronpedia

    INDEX
    Explanations

    effect

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     barrels
    -0.07
    -windows
    -0.07
     Masks
    -0.07
    .team
    -0.06
    according
    -0.06
    -dashboard
    -0.06
     lowers
    -0.06
     wagon
    -0.06
     pictures
    -0.06
    girls
    -0.06
    POSITIVE LOGITS
    	Z
    0.07
    obraz
    0.07
    �
    0.06
     prvním
    0.06
    ाभ
    0.06
    unsch
    0.06
    ]="
    0.06
    Qui
    0.06
    Ptr
    0.06
    LV
    0.06
    Act Density 0.055%

    No Known Activations