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

    Neuronpedia

    INDEX
    Explanations

    unknowingly doing something

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝑠
    2.04
     harassing
    1.96
    𝑢
    1.92
    мости
    1.89
    𝑚
    1.78
     едино
    1.78
    𝑑
    1.74
    بيقات
    1.73
     disturbing
    1.70
     سلسلے
    1.69
    POSITIVE LOGITS
    ar
    1.79
    k
    1.77
    uted
    1.74
    z
    1.68
    scheme
    1.65
    ay
    1.64
    f
    1.61
    项
    1.58
    acc
    1.57
    ネット
    1.52
    Act Density 0.001%

    No Known Activations