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

    Neuronpedia

    INDEX
    Explanations

    everySometimeslife

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     непосредственно
    0.93
     соответствующих
    0.87
    追加
    0.86
    는데요
    0.85
    较为
    0.84
     مشابه
    0.83
     例如
    0.83
    例如
    0.81
    ضاف
    0.80
     추가
    0.79
    POSITIVE LOGITS
     every
    1.86
     sometimes
    1.83
     Sometimes
    1.65
     Every
    1.65
     life
    1.59
    sometimes
    1.55
     everyone
    1.54
    every
    1.49
    Sometimes
    1.46
     loneliness
    1.42
    Act Density 0.331%

    No Known Activations