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

    Neuronpedia

    INDEX
    Explanations

    programming code

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sizeable
    0.61
    にて
    0.59
     ভীষণ
    0.59
    이랑
    0.57
     alakalı
    0.57
    もしくは
    0.57
     같이
    0.56
     हमको
    0.56
     Amongst
    0.55
     πως
    0.54
    POSITIVE LOGITS
    嗎
    0.55
     meningkatkan
    0.54
     বাঙ্গালীদের
    0.54
    ⁇
    0.52
     அமெ
    0.51
    <unused2221>
    0.50
    <eos>
    0.49
     ্
    0.49
    匤
    0.49
    蕞
    0.48
    Act Density 1.294%

    No Known Activations