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

    Neuronpedia

    INDEX
    Explanations

    base level or as a foundation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ین
    0.78
    ар
    0.72
    ю
    0.70
     уви
    0.68
     zacz
    0.67
    لي
    0.66
    я
    0.66
    пробу
    0.63
    ید
    0.63
     испыта
    0.63
    POSITIVE LOGITS
     Base
    1.44
     base
    1.29
    Base
    1.27
    base
    1.18
     for
    1.05
     BASE
    1.05
    BASE
    0.90
     बेस
    0.90
     Bases
    0.90
    ฐาน
    0.87
    Act Density 0.042%

    No Known Activations