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

    Neuronpedia

    INDEX
    Explanations

    attends to specific tokens marked as ** from various tokens marked as [[ ]]

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.23
    1:0.23
    2:0.17
    3:0.07
    4:0.07
    5:0.03
    6:0.06
    7:0.09
    Negative Logits
     uniqlo
    -0.46
    tôi
    -0.45
    */;
    -0.45
     greateſt
    -0.45
    endphp
    -0.45
     Shakspeare
    -0.45
     myſelf
    -0.45
    };*/
    -0.44
     ?</
    -0.44
    fromnode
    -0.43
    POSITIVE LOGITS
    '
    0.33
    <eos>
    0.32
     n
    0.31
    :
    0.31
    л
    0.31
    1
    0.31
    tr
    0.30
    ason
    0.29
    ann
    0.29
                        
    0.28
    Act Density 0.148%

    No Known Activations