INDEX
    Explanations

    making the world better

    The neuron activates strongly on optimistic calls to “make the world a better place,” i.e. uplifting, world-improvement slogans.

    New Auto-Interp
    Negative Logits
     قر
    -0.07
     Military
    -0.07
    declar
    -0.07
    ERROR
    -0.06
    _message
    -0.06
    ][:
    -0.06
    849
    -0.06
    _aux
    -0.06
    .Render
    -0.06
     cred
    -0.06
    POSITIVE LOGITS
     Lok
    0.06
    ToProps
    0.06
    IDEOS
    0.06
    うん
    0.06
    [contains
    0.06
     ülk
    0.06
    ừng
    0.06
    원의
    0.06
    ève
    0.06
    有些
    0.05
    Act Density 0.021%

    No Known Activations