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

    Neuronpedia

    INDEX
    Explanations

    say positive attributes

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     work
    0.70
     Work
    0.67
     работу
    0.65
     warm
    0.63
     lavoro
    0.62
     trabajo
    0.61
     работы
    0.61
    Work
    0.59
     Warm
    0.59
     trabalho
    0.59
    POSITIVE LOGITS
     fairly
    0.74
    Quiet
    0.47
     quiet
    0.46
     top
    0.43
     left
    0.42
    quiet
    0.42
    detail
    0.41
     Quiet
    0.40
    left
    0.39
     detail
    0.39
    Act Density 0.000%

    No Known Activations