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

    Neuronpedia

    INDEX
    Explanations

    attends to opening curly brackets from sequences of closing curly brackets

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.13
    1:0.18
    2:0.14
    3:0.09
    4:0.09
    5:0.06
    6:0.09
    7:0.19
    Negative Logits
    XmlAccessType
    -0.31
     AssemblyTitle
    -0.29
    Datuak
    -0.29
     resourceCulture
    -0.28
    millimeters
    -0.28
    kespeare
    -0.28
    inburgh
    -0.28
    ))){
    -0.28
    nestjs
    -0.28
    putInt
    -0.27
    POSITIVE LOGITS
    Hentet
    0.34
    Personensuche
    0.29
     externi
    0.29
    InjectAttribute
    0.27
     invokingState
    0.25
    вање
    0.25
    AxisAlignment
    0.25
     Moussa
    0.25
    BorderFactory
    0.24
    ậc
    0.24
    Act Density 0.057%

    No Known Activations