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

    Neuronpedia

    INDEX
    Explanations

    attends to the token "and" from other tokens separated by punctuation within a list or collection

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.17
    1:0.21
    2:0.14
    3:0.07
    4:0.08
    5:0.04
    6:0.07
    7:0.17
    Negative Logits
    HttpGet
    -0.20
     губ
    -0.20
    paw
    -0.20
     Funds
    -0.20
    phorbia
    -0.20
    nloa
    -0.20
    Entities
    -0.19
    neiden
    -0.19
    Funds
    -0.19
    geld
    -0.19
    POSITIVE LOGITS
    expandindo
    0.37
    новниш
    0.36
     réessayer
    0.34
    EDEFAULT
    0.33
    yntaxException
    0.33
    astéro
    0.33
    цезда
    0.33
     Extragalactic
    0.32
    actionMode
    0.32
    AutoScaleMode
    0.32
    Act Density 0.126%

    No Known Activations