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

    Neuronpedia

    INDEX
    Explanations

    correct output and accurate calculation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     diary
    0.35
    NameValuePair
    0.35
     diye
    0.34
     seized
    0.33
     apariencia
    0.33
     بیوی
    0.33
    arında
    0.32
     Aviv
    0.32
     کامل
    0.32
    itarian
    0.31
    POSITIVE LOGITS
    正确
    0.54
    正確
    0.54
     correctly
    0.52
     accurately
    0.50
    金额
    0.47
     calculates
    0.46
    金額
    0.46
     accurate
    0.46
     prawidł
    0.46
    正确的
    0.45
    Act Density 0.000%

    No Known Activations