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

    Neuronpedia

    INDEX
    Explanations

    review and recall information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     because
    0.43
     they
    0.41
     warning
    0.41
    安全
    0.38
    "
    0.38
    ᾽
    0.38
     being
    0.38
     not
    0.38
     which
    0.38
     silent
    0.37
    POSITIVE LOGITS
    缛
    0.46
    踴
    0.41
     เพิ่ม
    0.41
    gtrsim
    0.41
     Cliquez
    0.40
    鲷
    0.40
     مخطط
    0.40
     כבר
    0.39
    oljš
    0.39
     मिश्रण
    0.38
    Act Density 0.001%

    No Known Activations