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

    Neuronpedia

    INDEX
    Explanations

    attends to European tokens from related guideline tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.08
    1:0.10
    2:0.09
    3:0.12
    4:0.11
    5:0.06
    6:0.23
    7:0.17
    Negative Logits
     дописавши
    -0.31
    InjectMocks
    -0.28
     Brandenburg
    -0.26
    ęp
    -0.26
    も行
    -0.26
    literals
    -0.26
    ciuto
    -0.25
    addElement
    -0.25
    nand
    -0.25
    TargetApi
    -0.25
    POSITIVE LOGITS
    principalColumn
    0.41
     purpoſe
    0.35
     hunne
    0.35
    Personensuche
    0.34
     auffi
    0.34
     pleaſure
    0.34
     reaſon
    0.34
    ſelves
    0.33
    ſelf
    0.33
     juſ
    0.33
    Act Density 0.494%

    No Known Activations