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

    Neuronpedia

    INDEX
    Explanations

    Everything

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Something
    0.90
    Something
    0.90
     something
    0.84
     Someone
    0.72
     someone
    0.69
    Someone
    0.69
    something
    0.67
     somebody
    0.65
     Somebody
    0.65
     మారి
    0.62
    POSITIVE LOGITS
     but
    1.71
    but
    1.55
    But
    1.38
     But
    1.38
    但
    1.31
     लेकिन
    1.30
     nhưng
    1.30
     pero
    1.28
     maar
    1.25
     కానీ
    1.25
    Act Density 0.024%

    No Known Activations