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

    Neuronpedia

    INDEX
    Explanations

    legal and site status issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ।
    0.99
    you
    0.95
    。
    0.93
    AN
    0.89
    ORE
    0.86
    YOU
    0.83
    ט
    0.78
     anglais
    0.76
    IN
    0.73
    İ
    0.73
    POSITIVE LOGITS
     звез
    1.04
    and
    1.02
    ik
    1.00
    ↵
    0.95
    ?
    0.94
    ud
    0.91
     are
    0.90
    ни
    0.89
    a
    0.88
    の後
    0.88
    Act Density 3.672%

    No Known Activations