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

    Neuronpedia

    INDEX
    Explanations

    agreed upon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    дера
    0.70
    где
    0.68
     وكانت
    0.67
    दर्शिता
    0.65
    мышлен
    0.64
     tarik
    0.64
    まだまだ
    0.63
    ismatch
    0.63
    到底
    0.63
     এতদিন
    0.63
    POSITIVE LOGITS
     upon
    5.92
     Upon
    5.40
    Upon
    5.21
    upon
    5.13
     अपॉन
    2.29
     asupra
    2.07
     üzerine
    1.93
     عليه
    1.87
     върху
    1.79
     عليها
    1.53
    Act Density 0.231%

    No Known Activations