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

    Neuronpedia

    INDEX
    Explanations

    attends to digits from surrounding tokens related to legal and numerical contexts

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.11
    2:0.14
    3:0.19
    4:0.18
    5:0.16
    6:0.08
    7:0.03
    Negative Logits
    aarrggbb
    -0.45
     unknownFields
    -0.39
     ModelExpression
    -0.38
    DebuggerNonUser
    -0.35
     NSCoder
    -0.34
    曖昧さ回避
    -0.32
    LookAnd
    -0.31
    WithIOException
    -0.31
    ViewImports
    -0.31
     onAnimation
    -0.31
    POSITIVE LOGITS
     brancas
    0.30
    anueva
    0.28
     felizes
    0.26
     refractive
    0.25
     Partagez
    0.25
    méable
    0.25
     lumineux
    0.25
    Географија
    0.24
    őle
    0.24
    umenten
    0.24
    Act Density 0.091%

    No Known Activations