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

    Neuronpedia

    INDEX
    Explanations

    correct harmful behavior

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tuned
    0.46
     Raceway
    0.45
     tình
    0.42
     garage
    0.42
     ഇക്കാര
    0.42
    garage
    0.42
     transposed
    0.42
    ển
    0.41
    াণ
    0.41
     tuned
    0.41
    POSITIVE LOGITS
     Reprodu
    0.42
     Chop
    0.41
     Imit
    0.39
     pudieron
    0.39
    改
    0.38
    অথ
    0.37
    Reprodu
    0.37
     Cons
    0.36
    สมัย
    0.36
     Recogn
    0.35
    Act Density 0.006%

    No Known Activations