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

    Neuronpedia

    INDEX
    Explanations

    role will you play

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Sweep
    0.48
    ﺠ
    0.45
    кает
    0.44
     महिन्यात
    0.43
    Ꮺ
    0.42
    一个月
    0.42
     regularmente
    0.42
     trabalhando
    0.42
    ļu
    0.41
     Swipe
    0.41
    POSITIVE LOGITS
    an
    0.55
    com
    0.47
    th
    0.46
     bolted
    0.45
    re
    0.45
     shook
    0.45
    it
    0.45
     simplicity
    0.44
    ones
    0.44
     motions
    0.43
    Act Density 0.006%

    No Known Activations