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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens that indicate specific categories or items from sequences that refer to administrative terms or items associated with those categories

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.36
    1:0.22
    2:0.10
    3:0.06
    4:0.07
    5:0.02
    6:0.06
    7:0.07
    Negative Logits
    tor
    -0.18
    kozá
    -0.18
     BorderRadius
    -0.17
     riscos
    -0.17
     estekak
    -0.17
    STANDING
    -0.17
    kozás
    -0.17
    spent
    -0.17
    dés
    -0.17
     Schluss
    -0.16
    POSITIVE LOGITS
     ffilmiau
    0.34
    RectangleBorder
    0.32
    WireFormat
    0.30
     StringVar
    0.29
    ISupport
    0.29
    0.29
     AssemblyCulture
    0.28
    EndGlobalSection
    0.27
    AccessorTable
    0.26
    TagMode
    0.26
    Act Density 1.160%

    No Known Activations