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

    Neuronpedia

    INDEX
    Explanations

    software concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     $|
    1.08
     -
    1.03
     $-
    0.98
     |
    0.94
     \|
    0.93
     $-$
    0.92
     $=\
    0.90
     Membership
    0.90
     gg
    0.89
     $=$
    0.89
    POSITIVE LOGITS
     decompression
    0.77
    fadeIn
    0.73
    erdale
    0.71
    0.71
    តែ
    0.71
     липня
    0.70
    ersi
    0.70
     Salford
    0.69
    0.68
     болот
    0.67
    Act Density 0.103%

    No Known Activations