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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens marked as "lens" from tokens marked as "::"

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.08
    1:0.10
    2:0.11
    3:0.13
    4:0.10
    5:0.09
    6:0.18
    7:0.16
    Negative Logits
    DeclareMath
    -0.32
    htbp
    -0.31
    elseif
    -0.31
     Cik
    -0.31
    Dienstag
    -0.30
     Tapa
    -0.30
    ual
    -0.29
    相
    -0.29
     cordi
    -0.29
    UAL
    -0.29
    POSITIVE LOGITS
    tagext
    0.44
     AssemblyTitle
    0.40
    ConstraintMaker
    0.35
    Personendaten
    0.35
     partielle
    0.34
     المعيارى
    0.34
    coglie
    0.33
    udesta
    0.32
    ignements
    0.32
    TintMode
    0.31
    Act Density 0.155%

    No Known Activations