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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens related to links or URLs from tokens that denote subjects, such as names

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.06
    2:0.07
    3:0.05
    4:0.04
    5:0.02
    6:0.17
    7:0.49
    Negative Logits
     genom
    -0.23
    ://$
    -0.22
    
    -0.21
     ziyaret
    -0.20
     MASSACHUSETTS
    -0.20
    ز
    -0.20
    h
    -0.20
    ogonal
    -0.19
     compren
    -0.19
    minos
    -0.19
    POSITIVE LOGITS
    featureID
    0.43
    elemField
    0.35
    ьаж
    0.33
     propOrder
    0.33
    richTextPanel
    0.31
    ArrowToggle
    0.31
    picasso
    0.30
    contentLoaded
    0.30
    TypedDataSet
    0.29
    NOPQRST
    0.29
    Act Density 0.782%

    No Known Activations