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

    Neuronpedia

    INDEX
    Explanations

    pattern followed by specific words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Noise
    0.50
    Bye
    0.48
     obj
    0.45
     Trunk
    0.43
     মালিক
    0.42
    ু
    0.41
     Noise
    0.41
    ول
    0.41
    Bytes
    0.41
     caract
    0.40
    POSITIVE LOGITS
    and
    0.48
    ಗೊಂಡ
    0.46
    ଝ
    0.46
    independent
    0.45
     методом
    0.45
    peak
    0.44
    supplied
    0.44
    senior
    0.44
    are
    0.43
    reaching
    0.43
    Act Density 0.003%

    No Known Activations