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

    Neuronpedia

    INDEX
    Explanations

    SO with descriptive words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     использова
    0.79
    祭り
    0.77
     आल्सो
    0.76
    тся
    0.75
    ையும்
    0.75
    t
    0.75
    филлер
    0.75
    。)
    0.73
    사용
    0.70
    ।’
    0.70
    POSITIVE LOGITS
    -
    1.12
    {
    1.00
     are
    0.93
    é
    0.92
    ian
    0.90
    ice
    0.86
    are
    0.84
    ер
    0.84
    ia
    0.84
    ية
    0.83
    Act Density 0.022%

    No Known Activations