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

    Neuronpedia

    INDEX
    Explanations

    proposed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    可以
    1.43
     juist
    1.41
    1.41
    d
    1.38
     Alive
    1.36
     puissiez
    1.35
    その
    1.34
    ️⃣
    1.33
    般
    1.33
    ış
    1.30
    POSITIVE LOGITS
    ي
    2.56
    iation
    1.73
    লি
    1.70
     Interestingly
    1.62
    ون
    1.57
    대로
    1.55
    دت
    1.55
    י
    1.55
    ি
    1.55
     situés
    1.54
    Act Density 0.007%

    No Known Activations