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

    Neuronpedia

    INDEX
    Explanations

    tangible objects and items held or carried by characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     wink
    -0.17
     blink
    -0.15
    znik
    -0.15
     raz
    -0.15
    ilin
    -0.14
    anka
    -0.14
    line
    -0.14
    istant
    -0.14
    çīĻ
    -0.14
    inkle
    -0.14
    POSITIVE LOGITS
    held
    0.20
     held
    0.18
     Held
    0.17
    æĢĢ
    0.16
    ÑĨеп
    0.15
    porte
    0.15
    arme
    0.15
     carried
    0.15
    à¸Ńà¸ĩà¸Īาà¸ģ
    0.15
    ÏĨεÏģ
    0.14
    Act Density 0.142%

    No Known Activations