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

    Neuronpedia

    INDEX
    Explanations

    comparison of levels or degrees

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Mow
    0.47
     Hathaway
    0.42
     invers
    0.42
    更是
    0.41
     다섯
    0.40
     duas
    0.40
     Mā
    0.39
     inversion
    0.39
    🫣
    0.39
     여섯
    0.38
    POSITIVE LOGITS
     lesser
    1.05
     mindre
    0.90
     weaker
    0.88
     smaller
    0.88
     Lesser
    0.85
    Smaller
    0.85
    Secondary
    0.81
     Smaller
    0.80
    smaller
    0.80
     secondary
    0.77
    Act Density 0.497%

    No Known Activations