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

    Neuronpedia

    INDEX
    Explanations

    intensifier for descriptive words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     parfaitement
    0.93
     meilleure
    0.87
     migliori
    0.83
     mejores
    0.83
     tamamen
    0.82
    更好的
    0.82
     meilleures
    0.80
    更好
    0.77
     clearer
    0.76
     totalement
    0.75
    POSITIVE LOGITS
     large
    1.21
     small
    1.05
     high
    1.04
     tall
    0.99
     strong
    0.98
    large
    0.96
     loud
    0.96
     heavy
    0.96
     big
    0.92
     dense
    0.92
    Act Density 3.665%

    No Known Activations