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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens that form parts of programming expressions from arbitrarily named token expressions

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.10
    1:0.11
    2:0.12
    3:0.15
    4:0.11
    5:0.09
    6:0.16
    7:0.13
    Negative Logits
    UnsafeEnabled
    -0.38
    isDisplayed
    -0.26
    ъ
    -0.26
    IntoConstraints
    -0.24
     Kwan
    -0.23
    MLLoader
    -0.23
     Forder
    -0.23
    AsyncResult
    -0.22
     vain
    -0.22
     saw
    -0.22
    POSITIVE LOGITS
     للمعارف
    0.31
    AndEndTag
    0.29
    utilisons
    0.29
    cotch
    0.28
    rafat
    0.27
    awtextra
    0.27
     <>",
    0.27
    caloosa
    0.27
     ProtoMessage
    0.26
     gawas
    0.26
    Act Density 0.124%

    No Known Activations