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

    Neuronpedia

    INDEX
    Explanations

    加劇, 加上, 加冰, and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    আরো
    0.41
    क्स्ट
    0.41
    ണേ
    0.41
    ّر
    0.40
    تف
    0.39
    George
    0.38
    Formats
    0.37
    陬
    0.37
    抜き
    0.36
    ствие
    0.36
    POSITIVE LOGITS
    ⼊
    0.47
    持
    0.46
    速
    0.43
     mont
    0.42
    害者
    0.42
     atento
    0.41
     モン
    0.41
     hold
    0.41
     Sharpe
    0.40
     राजीव
    0.39
    Act Density 0.007%

    No Known Activations