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

    Neuronpedia

    INDEX
    Explanations

    attends to opening parentheses from later tokens expressing data types or variables

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.10
    2:0.08
    3:0.07
    4:0.05
    5:0.03
    6:0.08
    7:0.47
    Negative Logits
    ReusableCell
    -0.27
    ং
    -0.27
    Спољашње
    -0.25
    cinct
    -0.24
     للمعارف
    -0.24
     arka
    -0.24
    LabelTagHelper
    -0.24
    ங்க
    -0.24
     proced
    -0.23
    顺
    -0.23
    POSITIVE LOGITS
    toggler
    0.35
    fikasi
    0.33
    enschappelijke
    0.32
     насељу
    0.31
     تكبرها
    0.31
    Geographie
    0.31
    /*
    0.29
    σιμοποι
    0.29
    ِّف
    0.28
     Bourgoin
    0.28
    Act Density 0.072%

    No Known Activations