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

    Neuronpedia

    INDEX
    Explanations

    improve or better performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ১০
    1.23
    pa
    1.07
     pyar
    1.07
    ಬ್ಬಿಣ
    1.05
    AY
    1.05
     Atual
    1.04
    selves
    1.02
     fibras
    1.02
     branca
    1.02
    おります
    1.01
    POSITIVE LOGITS
    માં
    1.27
    િસ
    1.13
    га
    1.02
    ية
    1.02
    ो
    0.98
    െടുത്ത
    0.97
    ная
    0.97
     suited
    0.95
     যতটুকু
    0.94
    सी
    0.92
    Act Density 0.390%

    No Known Activations