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

    Neuronpedia

    INDEX
    Explanations

    attends to punctuation tokens from numerical or character sequences

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.08
    1:0.08
    2:0.16
    3:0.39
    4:0.09
    5:0.06
    6:0.05
    7:0.06
    Negative Logits
     aimer
    -0.29
    +:+
    -0.28
    vano
    -0.27
     favoring
    -0.26
    </em>
    -0.26
     Cual
    -0.25
     favored
    -0.25
    tof
    -0.24
    של
    -0.24
     herhangi
    -0.24
    POSITIVE LOGITS
     مشين
    0.41
    AndEndTag
    0.38
    AutoScaleMode
    0.37
    دانشنامهٔ
    0.36
     TextAppearance
    0.36
     BoxFit
    0.35
     '\\;'
    0.35
     cherchés
    0.35
    expandindo
    0.35
    kloped
    0.34
    Act Density 0.314%

    No Known Activations