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

    Neuronpedia

    INDEX
    Explanations

    attends to previously mentioned tokens from specific later tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.10
    2:0.07
    3:0.08
    4:0.45
    5:0.07
    6:0.06
    7:0.07
    Negative Logits
    RuleContext
    -0.35
     Numerade
    -0.34
    хьтан
    -0.32
    rowave
    -0.31
    windowFixed
    -0.30
    lanta
    -0.30
     صوتيه
    -0.30
    TintMode
    -0.29
    Coordonnées
    -0.29
    etCode
    -0.29
    POSITIVE LOGITS
     bianche
    0.36
     كومونز
    0.33
    jsxFileName
    0.30
    RequestMapping
    0.30
    спонд
    0.29
    GetMapping
    0.29
     handleMessage
    0.29
     spagno
    0.28
     IBOutlet
    0.28
     EClass
    0.27
    Act Density 0.587%

    No Known Activations