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

    Neuronpedia

    INDEX
    Explanations

    Fibonacci numbers, saving data

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ون
    0.80
    ل
    0.73
    و
    0.72
    ку
    0.66
    نا
    0.61
    ه
    0.61
    の
    0.61
    ين
    0.61
    лі
    0.59
    ის
    0.59
    POSITIVE LOGITS
    0.81
    ra
    0.52
     Перейти
    0.49
     С
    0.47
    .”
    0.46
    ating
    0.45
    ates
    0.44
     JPMorgan
    0.44
     Proposition
    0.44
     F
    0.43
    Act Density 2.122%

    No Known Activations