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

    Neuronpedia

    INDEX
    Explanations

    examples, explanations, research

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0
    0.43
    patient
    0.42
    psilon
    0.40
    asm
    0.40
     مشغول
    0.40
    forcing
    0.39
    igation
    0.39
    rystall
    0.39
     तू
    0.39
    riff
    0.39
    POSITIVE LOGITS
    🏘
    0.52
     abra
    0.50
    ה
    0.49
     ヴィンテージ
    0.47
    IE
    0.46
     כמו
    0.46
    કારે
    0.45
    zam
    0.45
     rada
    0.45
    ⠄
    0.45
    Act Density 0.002%

    No Known Activations