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

    Neuronpedia

    INDEX
    Explanations

    wp

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     shielding
    -0.07
     ensuing
    -0.07
     bonuses
    -0.07
    ocused
    -0.07
    equals
    -0.06
    Effect
    -0.06
    Card
    -0.06
    ，有
    -0.06
     Rectangle
    -0.06
    rias
    -0.06
    POSITIVE LOGITS
    '];↵
    0.07
     gent
    0.07
     �
    0.06
    voř
    0.06
    лик
    0.06
    !');↵
    0.06
    「你
    0.06
     Candidates
    0.06
     спіл
    0.06
    ’am
    0.06
    Act Density 0.002%

    No Known Activations