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

    Neuronpedia

    INDEX
    Explanations

    ethical and harmless principles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    大胆
    0.46
     hydrophobic
    0.46
     runtime
    0.43
    ู่
    0.42
     nightlife
    0.40
    bold
    0.40
    Runtime
    0.40
     bold
    0.39
     夜
    0.39
    Bold
    0.39
    POSITIVE LOGITS
     moral
    0.99
     altru
    0.95
     wholesome
    0.93
    Moral
    0.90
    善良
    0.89
     Moral
    0.89
     virtuous
    0.89
    moral
    0.85
     morals
    0.83
     ethical
    0.82
    Act Density 0.392%

    No Known Activations