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

    Neuronpedia

    INDEX
    Explanations

    butterfly effect, EU, compression

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     eğer
    0.48
     dacă
    0.43
    ជូន
    0.43
     if
    0.43
     diferença
    0.41
     dining
    0.41
     nếu
    0.41
     choix
    0.40
     multiplicative
    0.40
     moje
    0.40
    POSITIVE LOGITS
     הס
    0.48
     בת
    0.47
     பய
    0.47
     הר
    0.47
     הי
    0.46
    ני
    0.44
     అనేక
    0.44
    ๘
    0.44
     стро
    0.43
     архів
    0.43
    Act Density 0.007%

    No Known Activations