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

    Neuronpedia

    INDEX
    Explanations

    attends to the token "key" from tokens related to encryption and signing processes

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.10
    1:0.13
    2:0.11
    3:0.15
    4:0.13
    5:0.09
    6:0.12
    7:0.12
    Negative Logits
     surla
    -0.30
     nakalista
    -0.30
    jooq
    -0.28
    ReusableCell
    -0.26
    (/^
    -0.25
    icorn
    -0.24
    moveToFirst
    -0.24
     nonatomic
    -0.24
    IBar
    -0.23
    HomeAsUpEnabled
    -0.23
    POSITIVE LOGITS
     насељу
    0.32
     soprano
    0.26
     obstante
    0.25
    ınca
    0.25
    graphiques
    0.24
    หว่าง
    0.24
     gbc
    0.23
     însă
    0.23
    それにしても
    0.22
    endpush
    0.22
    Act Density 0.154%

    No Known Activations