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

    Neuronpedia

    INDEX
    Explanations

    pop in or out

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Ти
    2.02
     т
    1.96
    Бе
    1.95
     Бе
    1.91
    ก
    1.90
    되었
    1.88
    ج
    1.88
    г
    1.88
    ীন্দ্র
    1.84
    마
    1.84
    POSITIVE LOGITS
    ulating
    1.84
    ть
    1.80
    ters
    1.74
    ulated
    1.70
    tery
    1.65
    che
    1.64
    brett
    1.57
    threads
    1.54
    sticks
    1.51
    waffe
    1.48
    Act Density 0.004%

    No Known Activations