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

    Neuronpedia

    INDEX
    Explanations

    was a

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Very
    1.17
     Very
    1.12
     very
    1.02
    很不
    0.96
    very
    0.96
     veldig
    0.95
    Очень
    0.94
    很是
    0.94
     очень
    0.94
     sehr
    0.92
    POSITIVE LOGITS
     ذلك
    0.85
     ciò
    0.78
     everything
    0.76
     elegance
    0.76
     ấy
    0.73
     đó
    0.72
     therein
    0.71
     sempurna
    0.68
     ਨੇ
    0.68
     brilliance
    0.68
    Act Density 0.165%

    No Known Activations