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

    Neuronpedia

    INDEX
    Explanations

    attention

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     rabatt
    -0.09
     visant
    -0.09
     accessing
    -0.08
     ഖ
    -0.08
     जुन
    -0.08
     ideaal
    -0.08
    punkt
    -0.08
     ve
    -0.07
     ಶಿವ
    -0.07
     jambes
    -0.07
    POSITIVE LOGITS
     Mr
    0.08
     prestar
    0.08
     Dar
    0.08
    Dar
    0.08
     atenção
    0.08
     Display
    0.08
     Updated
    0.07
    Mr
    0.07
     dar
    0.07
     fría
    0.07
    Act Density 0.001%

    No Known Activations