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

    Neuronpedia

    INDEX
    Explanations

    request violates safety severe ways

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    aduras
    0.40
     bargaining
    0.40
    rz
    0.39
     altas
    0.38
     adaptive
    0.38
    č
    0.38
     Preventive
    0.38
     robotic
    0.37
     Microbial
    0.37
     adaptation
    0.37
    POSITIVE LOGITS
    有两个
    0.47
     fundamentales
    0.46
     அடிப்பட
    0.38
     utama
    0.37
     parametri
    0.37
     aren
    0.36
     DIRECTIONS
    0.35
     beyond
    0.35
     weighty
    0.35
    beyond
    0.35
    Act Density 0.030%

    No Known Activations