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

    Neuronpedia

    INDEX
    Explanations

    buttons and tables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    at
    0.64
    га
    0.62
    ប
    0.62
    écou
    0.59
    kBtu
    0.57
    g
    0.57
    لال
    0.57
    ках
    0.54
    ANCIS
    0.54
    دا
    0.53
    POSITIVE LOGITS
    ര്‍ക്കും
    0.65
     Overwatch
    0.59
     Destroyer
    0.59
     wary
    0.58
    ര്‍
    0.57
    ยัง
    0.57
     Poincaré
    0.56
     Commander
    0.55
     жизнь
    0.55
     Nazis
    0.55
    Act Density 0.004%

    No Known Activations