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

    Neuronpedia

    INDEX
    Explanations

    opening curly brace `{`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     of
    -1.80
     to
    -1.73
     the
    -1.70
     was
    -1.55
     recent
    -1.52
    </em>
    -1.44
     a
    -1.41
    。"
    -1.36
     an
    -1.34
    </h3>
    -1.34
    POSITIVE LOGITS
    rugada
    1.44
    quiries
    1.42
     Tutti
    1.41
     aasta
    1.41
    МІ
    1.38
    béco
    1.38
    Sabato
    1.38
     TAMBIÉN
    1.35
    不過
    1.35
    even
    1.35
    Act Density 0.017%

    No Known Activations