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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens containing "max" from closing brackets

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.11
    2:0.11
    3:0.12
    4:0.12
    5:0.10
    6:0.15
    7:0.16
    Negative Logits
    AndEndTag
    -0.39
     كومونز
    -0.37
    }))
    -0.33
     tartalomajánló
    -0.30
    gridx
    -0.29
    AsUp
    -0.29
    HasAnnotation
    -0.29
    queryInterface
    -0.28
    IVEREF
    -0.28
    GEBURTSDATUM
    -0.28
    POSITIVE LOGITS
     Houſe
    0.34
     houſe
    0.31
    neſs
    0.30
     Jefus
    0.30
     juſt
    0.29
     potentially
    0.29
    ISupport
    0.28
     Majefty
    0.28
     ſtate
    0.28
    jkl
    0.27
    Act Density 0.001%

    No Known Activations