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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens with asterisks from tokens with square brackets

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.13
    1:0.17
    2:0.14
    3:0.10
    4:0.15
    5:0.05
    6:0.07
    7:0.16
    Negative Logits
    IBOutlet
    -0.27
     colspan
    -0.27
    互
    -0.26
    umus
    -0.26
    migiano
    -0.26
    ↵
    -0.25
    ISupport
    -0.25
    NET
    -0.24
    anu
    -0.24
    CloseOperation
    -0.24
    POSITIVE LOGITS
     ujednoznacz
    0.48
    Hochspringen
    0.42
     bezeichneter
    0.41
    elry
    0.39
    UnsafeEnabled
    0.39
     zoude
    0.36
     auroit
    0.36
     feroit
    0.35
     universel
    0.35
     ainfi
    0.35
    Act Density 0.306%

    No Known Activations