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

    Neuronpedia

    INDEX
    Explanations

    marathon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    N
    0.43
     carpentry
    0.41
    SK
    0.40
    puppet
    0.40
    ’
    0.39
    สร
    0.37
    indicating
    0.37
     Adriatic
    0.37
    o
    0.37
    рова
    0.37
    POSITIVE LOGITS
     marathon
    0.66
     triathlon
    0.64
     runners
    0.61
    🏃
    0.55
     Runner
    0.54
     runner
    0.54
     Marathon
    0.52
     Runners
    0.49
    跑步
    0.49
    thon
    0.49
    Act Density 0.007%

    No Known Activations