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

    Neuronpedia

    INDEX
    Explanations

    academic terms and programming code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Bayesian
    1.02
     Environments
    1.00
     다양한
    0.99
     환경
    0.98
     낮은
    0.98
     쉽게
    0.97
     얼마나
    0.95
     heterogeneity
    0.92
     미국
    0.91
     macOS
    0.91
    POSITIVE LOGITS
    countdown
    0.79
    except
    0.74
    ;//
    0.68
    invited
    0.68
    ，
    0.67
    new
    0.67
    </div>
    0.67
    other
    0.66
    including
    0.66
    wi
    0.65
    Act Density 0.813%

    No Known Activations