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

    Neuronpedia

    INDEX
    Explanations

    official communication/announcements

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Bermuda
    -0.08
     free
    -0.08
    あり
    -0.08
    ，是
    -0.08
     afinal
    -0.08
     ane
    -0.07
    目です
    -0.07
    كرة
    -0.07
     fria
    -0.07
     protože
    -0.07
    POSITIVE LOGITS
    马会
    0.11
     chooses
    0.11
    ોએ
    0.10
     learns
    0.10
     сможет
    0.10
     argues
    0.10
    ողները
    0.10
     decides
    0.10
     چاروا
    0.10
    数据显示
    0.10
    Act Density 0.556%

    No Known Activations