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

    Neuronpedia

    INDEX
    Explanations

    potential and promising applications

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     debes
    0.47
     usually
    0.46
    總是
    0.46
     Usually
    0.44
     Unless
    0.44
     scegli
    0.43
     ไม่มี
    0.43
     morate
    0.43
    選び
    0.42
     завжди
    0.42
    POSITIVE LOGITS
     potentially
    1.34
     potencialmente
    1.30
     promising
    1.20
    potentially
    1.20
     потенциа
    1.16
    有望
    1.14
     potencial
    1.11
     promise
    1.08
     potential
    1.06
     promises
    1.05
    Act Density 0.016%

    No Known Activations