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

    Neuronpedia

    INDEX
    Explanations

    auto-rickshaws

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    し
    0.64
    <0x0D>
    0.62
    ف
    0.59
    i
    0.57
    い
    0.54
     hath
    0.51
     whiche
    0.51
     Aron
    0.51
    yal
    0.51
    '.$
    0.50
    POSITIVE LOGITS
     ಏಕೆ
    0.59
    в
    0.59
    ""
    0.55
     મળશે
    0.54
     ôm
    0.53
     ऑप्ट
    0.53
     রামগতি
    0.52
     स्टार्ट
    0.52
     इसलिए
    0.51
     জানান
    0.51
    Act Density 0.001%

    No Known Activations