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

    Neuronpedia

    INDEX
    Explanations

    terminates recursion or decides where

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    en
    0.38
    و
    0.34
    一个
    0.27
    at
    0.27
    灣
    0.26
    만원
    0.26
    vem
    0.25
    一个是
    0.24
     alten
    0.24
    什么是
    0.24
    POSITIVE LOGITS
    вання
    0.27
     જણાવ
    0.26
    .|__
    0.25
     дели
    0.24
    .[/
    0.24
    𝑒
    0.24
     Chom
    0.24
    ľov
    0.23
     godfather
    0.23
    \{\
    0.23
    Act Density 0.800%

    No Known Activations