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

    Neuronpedia

    INDEX
    Explanations

    questions or requests to summarize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ف
    0.54
    РИ
    0.54
    itze
    0.52
    (!)
    0.50
    雦
    0.48
    abble
    0.48
    Ма
    0.46
    بق
    0.45
    但
    0.45
    layoutWidget
    0.45
    POSITIVE LOGITS
    ？
    0.88
    ？）
    0.75
    ?).
    0.69
     আদৌ
    0.68
    ?),
    0.68
     müşter
    0.68
    ؟
    0.64
    ă
    0.64
     açı
    0.63
    ů
    0.61
    Act Density 2.135%

    No Known Activations