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

    Neuronpedia

    INDEX
    Explanations

    there's a

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     są
    0.96
     jsou
    0.94
     são
    0.88
     are
    0.88
     ovat
    0.82
    都是
    0.80
     janë
    0.78
     sont
    0.77
     هستند
    0.77
     vannak
    0.77
    POSITIVE LOGITS
     lies
    0.42
     goes
    0.40
    有一
    0.35
     was
    0.34
    няется
    0.34
     stands
    0.33
     Goes
    0.33
     heps
    0.32
     sits
    0.32
     comes
    0.32
    Act Density 0.098%

    No Known Activations