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

    Neuronpedia

    INDEX
    Explanations

    good answer descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    т
    0.58
    t
    0.51
    硒
    0.48
    Experiment
    0.47
    ா
    0.47
    гән
    0.45
    itação
    0.45
     карка
    0.42
    Brooks
    0.41
    iénd
    0.41
    POSITIVE LOGITS
     velmi
    0.50
     dvou
    0.48
     bagels
    0.47
     bastante
    0.46
     muito
    0.45
     Muito
    0.44
     wygląda
    0.44
     molto
    0.44
     bonnes
    0.44
    不错
    0.43
    Act Density 0.003%

    No Known Activations