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

    Neuronpedia

    INDEX
    Explanations

    outperform

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     }{\
    0.49
     berguna
    0.48
     впли
    0.48
    變成
    0.46
     gedacht
    0.45
     diterapkan
    0.44
     anuv
    0.44
    ීර
    0.44
     fonctionnement
    0.44
    变成
    0.44
    POSITIVE LOGITS
     consistently
    1.25
     outperform
    1.21
     outperformed
    1.18
     topping
    1.14
     outperforms
    1.11
     scored
    1.09
     scoring
    1.04
     topped
    1.04
     edged
    1.01
     surpass
    0.98
    Act Density 0.256%

    No Known Activations