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

    Neuronpedia

    INDEX
    Explanations

    positive descriptions and qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .");
    0.84
    <eos>
    0.84
     означает
    0.82
    "};
    0.82
    ");
    0.81
    ائية
    0.80
    "));
    0.80
    "/><
    0.80
     ہے۔
    0.79
    "];
    0.78
    POSITIVE LOGITS
     didnt
    0.98
     needing
    0.95
     wife
    0.95
     esposa
    0.89
     plenty
    0.89
     flew
    0.87
     wives
    0.87
     hadde
    0.87
     havde
    0.86
     same
    0.86
    Act Density 0.002%

    No Known Activations