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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens associated with private events from tokens related to public events

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.12
    1:0.16
    2:0.14
    3:0.13
    4:0.09
    5:0.03
    6:0.11
    7:0.18
    Negative Logits
    ieteur
    -0.33
    awtextra
    -0.30
    apnews
    -0.29
    GEBURTSDATUM
    -0.28
    hline
    -0.28
    roek
    -0.25
    lotten
    -0.25
    ulier
    -0.25
     schizophren
    -0.25
    ofür
    -0.25
    POSITIVE LOGITS
    OrBuilder
    0.41
    RuleContext
    0.37
    AndEndTag
    0.36
    ViewImports
    0.35
     francès
    0.35
    Kanpo
    0.34
    SerializedSize
    0.34
    يح
    0.34
     חיצוניים
    0.33
    zeitig
    0.33
    Act Density 0.005%

    No Known Activations