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

    Neuronpedia

    INDEX
    Explanations

    digits after periods or colons

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     مہارت
    0.35
    doesn
    0.34
     nazy
    0.34
    were
    0.34
     уж
    0.33
    पाई
    0.32
    currentToken
    0.32
    fileName
    0.31
    ೀರ್
    0.31
    ARON
    0.31
    POSITIVE LOGITS
    ‐
    0.37
     se
    0.36
    ج
    0.35
    5
    0.35
    ز
    0.35
    发
    0.34
    ↵
    0.33
    ات
    0.33
     half
    0.33
    </h2>
    0.32
    Act Density 0.063%

    No Known Activations