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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens that are followed by commas to tokens that are separated by commas

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.10
    1:0.12
    2:0.10
    3:0.13
    4:0.14
    5:0.10
    6:0.11
    7:0.17
    Negative Logits
    opla
    -0.34
     Moreira
    -0.28
    躇
    -0.28
    OP
    -0.28
    通
    -0.27
    Op
    -0.27
    tetten
    -0.27
    ং
    -0.26
     ellers
    -0.26
     dokon
    -0.26
    POSITIVE LOGITS
     للاسماء
    0.41
     pageContext
    0.37
    SBATCH
    0.37
     незавершена
    0.34
    +#+#
    0.34
    InSection
    0.34
     ostavi
    0.33
    addPreferredGap
    0.33
    DockStyle
    0.33
     '\\;'
    0.32
    Act Density 1.160%

    No Known Activations