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

    Neuronpedia

    INDEX
    Explanations

    generally harmless

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     அடிப்பட
    0.47
    ೈ
    0.46
    亼
    0.45
    ாவில்
    0.44
    تاریخ
    0.43
    فاف
    0.43
     Behold
    0.43
     seduced
    0.43
    摘
    0.42
    复杂的
    0.42
    POSITIVE LOGITS
     उसमें
    0.40
    Activity
    0.38
     proceedings
    0.38
     idiot
    0.38
     gmin
    0.38
    ActionButton
    0.37
    vdd
    0.37
     udział
    0.37
    }_
    0.37
    {-#
    0.37
    Act Density 0.013%

    No Known Activations