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

    Neuronpedia

    INDEX
    Explanations

    helpful and harmless

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    玮
    0.75
    вит
    0.75
    いただきます
    0.72
    imu
    0.71
     apuesta
    0.70
    ত্তা
    0.70
    ся
    0.70
    сол
    0.68
    рти
    0.68
    త్తు
    0.68
    POSITIVE LOGITS
     Hinweise
    0.83
     helpful
    0.77
    प्टन
    0.75
     Useful
    0.72
     useful
    0.72
     hilfre
    0.70
     জী
    0.68
    diag
    0.68
     foolish
    0.67
    useful
    0.67
    Act Density 0.106%

    No Known Activations