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

    Neuronpedia

    INDEX
    Explanations

    new and existing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     را
    0.30
    𝙜
    0.28
    ىسى
    0.28
    ﺮ
    0.27
     melindungi
    0.27
    akukan
    0.26
    را
    0.26
     থেকে
    0.26
     ा
    0.26
     ぐらい
    0.26
    POSITIVE LOGITS
     모든
    0.44
    その
    0.44
     새로운
    0.42
     그
    0.40
     처음
    0.39
    特に
    0.35
    新たに
    0.35
     특히
    0.35
     가장
    0.34
     반드시
    0.34
    Act Density 0.013%

    No Known Activations