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

    Neuronpedia

    INDEX
    Explanations

    crystal words

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     garn
    0.40
     escucha
    0.39
     Flo
    0.38
     Garn
    0.38
    Quartz
    0.38
     longiore
    0.38
     Quartz
    0.38
    Zinc
    0.37
    脆弱
    0.36
    obe
    0.35
    POSITIVE LOGITS
    Kv
    0.47
    Sf
    0.45
     cen
    0.45
     Cen
    0.43
     sf
    0.42
    يدات
    0.41
     kis
    0.41
     Bergen
    0.41
     Бер
    0.40
     CEN
    0.40
    Act Density 0.000%

    No Known Activations