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

    Neuronpedia

    INDEX
    Explanations

    instances of manipulation, particularly related to influence and control over others or processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.02
    2:0.05
    3:0.05
    4:0.13
    5:0.03
    6:0.03
    7:0.38
    8:0.05
    9:0.04
    10:0.09
    11:0.07
    Negative Logits
    chwitz
    -1.98
    alach
    -1.73
    ukemia
    -1.70
    pleted
    -1.70
    alty
    -1.69
    inguished
    -1.67
    emetery
    -1.66
    grave
    -1.61
     Cemetery
    -1.61
    stadt
    -1.58
    POSITIVE LOGITS
     levers
    2.15
     influence
    1.74
     ruler
    1.68
     curve
    1.66
     perceptions
    1.64
     grip
    1.61
     wedge
    1.59
     impulses
    1.59
     proxies
    1.58
     shape
    1.56
    Act Density 0.007%

    No Known Activations