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

    Neuronpedia

    INDEX
    Explanations

    Gemma

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     및
    0.65
     एवं
    0.64
     первич
    0.64
     formación
    0.63
     трансформа
    0.63
     apoio
    0.63
     homenagem
    0.63
     construcción
    0.62
     neoc
    0.62
     bibendum
    0.62
    POSITIVE LOGITS
     χρει
    1.01
     είναι
    0.97
     πρέπει
    0.82
     μπορεί
    0.82
     έχ
    0.80
     δεν
    0.71
     трябва
    0.68
     έχει
    0.67
     κάν
    0.66
     vreau
    0.65
    Act Density 0.005%

    No Known Activations