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

    Neuronpedia

    INDEX
    Explanations

    attends to positive political action tokens from negative broader discussion tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.10
    1:0.15
    2:0.13
    3:0.10
    4:0.12
    5:0.06
    6:0.11
    7:0.18
    Negative Logits
    Réalisation
    -0.22
    Ầ
    -0.21
    wnd
    -0.21
     Nó
    -0.20
     Shetty
    -0.20
    owo
    -0.20
    sighted
    -0.20
    Cô
    -0.20
    skraft
    -0.20
    obé
    -0.20
    POSITIVE LOGITS
    BeginContext
    0.40
    Personendaten
    0.37
     AssemblyCulture
    0.35
    EDEFAULT
    0.35
    SharedDtor
    0.34
     autorytatywna
    0.34
    Искәрмәләр
    0.32
    oredCriteria
    0.32
     HasFactory
    0.31
     المعيارى
    0.31
    Act Density 0.139%

    No Known Activations