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

    Neuronpedia

    INDEX
    Explanations

    complex/complicated

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     but
    0.95
    but
    0.90
     if
    0.85
     because
    0.83
     But
    0.77
     aren
    0.76
     Pei
    0.76
     दैट
    0.76
    pero
    0.75
    liono
    0.75
    POSITIVE LOGITS
     studie
    0.90
     con
    0.89
     ఆధ్యాత్మిక
    0.88
     Studi
    0.86
     studi
    0.85
     lớn
    0.83
     za
    0.82
    Study
    0.82
     सामाजिक
    0.80
    करिता
    0.79
    Act Density 0.001%

    No Known Activations