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

    Neuronpedia

    INDEX
    Explanations

    ever since

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    1.50
     завжди
    1.44
     всегда
    1.39
     siempre
    1.38
     always
    1.33
     Always
    1.30
    いる
    1.27
    Always
    1.26
     zawsze
    1.23
     sempre
    1.22
    POSITIVE LOGITS
    greens
    1.87
    lasting
    1.74
    glades
    1.62
    ../../
    1.28
    rrr
    1.26
    rrrr
    1.21
    grande
    1.18
    rr
    1.17
    رر
    1.17
    Кроме
    1.15
    Act Density 0.008%

    No Known Activations