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

    Neuronpedia

    INDEX
    Explanations

    gets a lot of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    perfect
    1.01
     perfect
    1.01
     perfectly
    0.99
    Perfect
    0.94
     Perfect
    0.92
     Perfectly
    0.87
     सर्वोत्तम
    0.87
    uitively
    0.82
     totally
    0.79
     entirely
    0.79
    POSITIVE LOGITS
    かなり
    1.11
     상당
    0.99
     bastante
    0.99
    不少
    0.99
     khá
    0.98
     firepower
    0.95
     banyak
    0.95
     alot
    0.94
     많이
    0.94
     ciddi
    0.92
    Act Density 0.258%

    No Known Activations