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

    Neuronpedia

    INDEX
    Explanations

    attends to the last column value from earlier rows paired with subsequent tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.08
    2:0.10
    3:0.09
    4:0.10
    5:0.07
    6:0.26
    7:0.20
    Negative Logits
    gegangen
    -0.29
    ",&
    -0.28
    /"+
    -0.27
    ShowDialog
    -0.25
    /'+
    -0.25
    ?),
    -0.25
    EndInit
    -0.25
    lt
    -0.24
    peth
    -0.24
    enci
    -0.24
    POSITIVE LOGITS
     الرياضيه
    0.43
     sublimation
    0.39
     BeautifulSoup
    0.38
     Parthen
    0.38
    +:+
    0.35
    mặt
    0.35
     Upan
    0.35
     Infórmanos
    0.35
     Administrativna
    0.34
    ​—
    0.34
    Act Density 0.032%

    No Known Activations