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

    Neuronpedia

    INDEX
    Explanations

    bolded phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    掐
    0.72
     Gosudarstvennyj
    0.68
     disallowed
    0.67
     こんな
    0.66
     circumstantial
    0.65
     仕様
    0.65
    वरील
    0.65
     ఇచ్చిన
    0.65
     スタッドレスタイヤ
    0.64
     manipulated
    0.64
    POSITIVE LOGITS
     Sky
    1.65
     Quantum
    1.63
     Nova
    1.63
     Blue
    1.62
     Thrive
    1.61
     Smart
    1.58
     Flex
    1.55
     Harmony
    1.54
     Horizon
    1.53
     Aqua
    1.53
    Act Density 0.660%

    No Known Activations