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

    Neuronpedia

    INDEX
    Explanations

    resembles or is like

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ги
    1.67
    𝐠
    1.60
    ز
    1.60
    𝐝
    1.55
    ৃত্ব
    1.55
    𝐧
    1.54
    𝐄
    1.50
     Inoltre
    1.48
    𝐞
    1.48
    然後
    1.41
    POSITIVE LOGITS
    ingly
    1.94
    不像
    1.70
    neſs
    1.67
    样的
    1.53
    minded
    1.53
    achute
    1.51
    ी
    1.50
     څنګه
    1.50
    istically
    1.48
    ॉम
    1.45
    Act Density 0.096%

    No Known Activations