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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens indicating positive performance or ability, from both direct assessments of players' gameplay and references to their performances in various contexts

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.17
    1:0.21
    2:0.15
    3:0.08
    4:0.08
    5:0.03
    6:0.06
    7:0.18
    Negative Logits
     typelib
    -0.34
    contentLoaded
    -0.33
    MigrationBuilder
    -0.30
     aveug
    -0.29
    IsMutable
    -0.29
    rrggbb
    -0.28
    PropertyChanging
    -0.28
    protoimpl
    -0.27
     تضيفلها
    -0.26
    WebVitals
    -0.25
    POSITIVE LOGITS
     Audiodateien
    0.30
    iastes
    0.24
    </h5>
    0.24
     Lidé
    0.23
    ATIONS
    0.23
    Ctx
    0.23
    jss
    0.22
    Inne
    0.22
    Sams
    0.22
     Ordin
    0.22
    Act Density 0.166%

    No Known Activations