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

    Neuronpedia

    INDEX
    Explanations

    edges, sides, bottoms, tops

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mirip
    0.70
     serupa
    0.67
     ক্ষমা
    0.64
    回事
    0.63
     çağ
    0.62
    同样的
    0.62
    ลิง
    0.61
     hukuk
    0.61
     fooling
    0.61
    प्लेक्स
    0.60
    POSITIVE LOGITS
     edges
    2.47
     sides
    2.24
     bottom
    2.13
     edge
    2.00
    edges
    1.92
     corners
    1.86
     ends
    1.79
    sides
    1.79
    bottom
    1.78
     bottoms
    1.76
    Act Density 0.216%

    No Known Activations