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

    Neuronpedia

    INDEX
    Explanations

    phrases related to health and sickness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.04
    2:0.13
    3:0.04
    4:0.04
    5:0.05
    6:0.23
    7:0.04
    8:0.03
    9:0.24
    10:0.02
    11:0.02
    Negative Logits
     Zhou
    -3.61
    ulia
    -3.57
    imble
    -3.52
    ragon
    -3.36
    arov
    -3.32
    ixel
    -3.32
    VB
    -3.30
     Decoder
    -3.27
    iframe
    -3.26
     Zimmerman
    -3.23
    POSITIVE LOGITS
     sick
    7.88
     Sick
    7.82
     sickness
    7.67
     illness
    5.92
     illnesses
    5.66
    health
    4.22
    Health
    4.13
     coughing
    4.11
     cured
    4.06
     health
    4.02
    Act Density 0.003%

    No Known Activations