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

    Neuronpedia

    INDEX
    Explanations

    attends to numeric tokens from adjacent parentheses or brackets

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.13
    1:0.12
    2:0.09
    3:0.06
    4:0.07
    5:0.04
    6:0.13
    7:0.32
    Negative Logits
     '\\;'
    -0.63
    EndGlobalSection
    -0.58
    CloseOperation
    -0.58
    NameInMap
    -0.55
     snippetHide
    -0.54
    Bioaccumulative
    -0.52
    aarrggbb
    -0.51
    httphttps
    -0.51
     kasarigan
    -0.50
     للاسماء
    -0.49
    POSITIVE LOGITS
    and
    0.33
    和
    0.33
    by
    0.33
    即
    0.33
    OR
    0.32
    ผ
    0.31
    ora
    0.31
    Il
    0.31
    the
    0.30
    2
    0.30
    Act Density 0.048%

    No Known Activations