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

    Neuronpedia

    INDEX
    Explanations

    organizing tasks and creation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     তিন
    0.45
    镗
    0.43
     تین
    0.42
     बिज़
    0.42
    0.42
     በጥ
    0.42
     ત્રણ
    0.41
     tři
    0.41
    GUNDABAD
    0.39
    鉀
    0.39
    POSITIVE LOGITS
    <0xBF>
    0.48
    9
    0.45
     or
    0.43
     nine
    0.38
     final
    0.36
    util
    0.35
     resulted
    0.35
     action
    0.34
     Ninth
    0.34
    /
    0.33
    Act Density 0.011%

    No Known Activations