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

    Neuronpedia

    INDEX
    Explanations

    Arabic conversational phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     تو
    0.87
     تُ
    0.87
     تضم
    0.86
     خلال
    0.86
    ت
    0.84
    ًا
    0.83
     ت
    0.81
     تم
    0.80
    ﺗ
    0.79
     مجموعة
    0.76
    POSITIVE LOGITS
     Beautiful
    0.79
     Enter
    0.78
     beautiful
    0.75
     Very
    0.75
     plek
    0.74
     frumo
    0.72
     vostri
    0.72
     Always
    0.70
     gentlemen
    0.70
     Sweet
    0.69
    Act Density 0.004%

    No Known Activations