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

    Neuronpedia

    INDEX
    Explanations

    highlighting existing capabilities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    auten
    0.55
    營
    0.50
    ↵
    0.49
    ique
    0.48
     walls
    0.44
    cą
    0.43
    oxo
    0.42
    llis
    0.41
    rát
    0.41
    ical
    0.40
    POSITIVE LOGITS
    ‌پ
    0.52
     Wege
    0.45
     Einige
    0.44
     upped
    0.44
     hilfre
    0.44
     tweaked
    0.43
     denkt
    0.43
     zahlreiche
    0.42
     nějak
    0.42
     erste
    0.42
    Act Density 0.001%

    No Known Activations