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

    Neuronpedia

    INDEX
    Explanations

    empty string

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Limits
    0.62
    եց
    0.62
     Portfolio
    0.61
    ～
    0.61
     مقر
    0.60
     ～
    0.60
     Mouth
    0.60
     ​​
    0.59
     }));
    0.58
     J
    0.57
    POSITIVE LOGITS
     Stal
    0.73
     الاشتراك
    0.68
     chom
    0.67
    抜
    0.65
    皙
    0.65
     alcançar
    0.65
     покры
    0.65
     cubrir
    0.65
     nim
    0.64
     nue
    0.64
    Act Density 0.001%

    No Known Activations