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

    Neuronpedia

    INDEX
    Explanations

    won / champion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     שהוא
    0.57
     vorhanden
    0.57
     soggetti
    0.57
     realmente
    0.56
     algunos
    0.56
     بعض
    0.55
     vorhand
    0.55
    が含ま
    0.55
     meditative
    0.55
     tiresome
    0.55
    POSITIVE LOGITS
     στο
    0.62
     on
    0.57
     for
    0.55
     спустя
    0.55
     στην
    0.53
     على
    0.53
     για
    0.52
     στον
    0.52
     فى
    0.50
     with
    0.50
    Act Density 0.001%

    No Known Activations