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

    Neuronpedia

    INDEX
    Explanations

    Final answer marker

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     brug
    0.78
     paves
    0.76
    ವರೆಗೆ
    0.73
    works
    0.73
    parking
    0.70
     উভয়ের
    0.70
     alike
    0.70
     canadian
    0.69
    yo
    0.69
     pendants
    0.68
    POSITIVE LOGITS
     Era
    0.86
    वास
    0.83
     Turns
    0.80
     Smell
    0.76
     Перед
    0.75
     jk
    0.74
    ερ
    0.74
     Surprisingly
    0.73
     Tricks
    0.73
    ల్య
    0.72
    Act Density 0.018%

    No Known Activations