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

    Neuronpedia

    INDEX
    Explanations

    morally wrong scenarios

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     depuis
    -0.08
    uted
    -0.07
    inut
    -0.07
     DATA
    -0.07
     confort
    -0.07
    -0.07
    centration
    -0.07
     জানান
    -0.07
    'heure
    -0.07
    geen
    -0.07
    POSITIVE LOGITS
     ataque
    0.09
     unnecessarily
    0.09
    偷
    0.08
    .alloc
    0.08
     করলে
    0.08
    攻击
    0.08
     dishonest
    0.08
     disrespect
    0.08
     зараж
    0.08
    thetho
    0.08
    Act Density 0.005%

    No Known Activations