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

    Neuronpedia

    INDEX
    Explanations

    harass or bully someone

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     against
    1.37
     Against
    1.20
    against
    1.20
     AGAINST
    1.18
    Against
    1.17
    对
    1.07
     проти
    1.01
     versus
    0.96
     ضد
    0.94
     contre
    0.93
    POSITIVE LOGITS
     merciless
    0.85
     angrily
    0.83
     Attempt
    0.81
    whom
    0.81
     provoca
    0.79
     whom
    0.79
    Attempt
    0.78
     avenge
    0.78
     provoquer
    0.77
     brutal
    0.77
    Act Density 0.132%

    No Known Activations