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

    Neuronpedia

    INDEX
    Explanations

    handling harmful content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     =$
    0.48
     eus
    0.47
    Id
    0.47
    É
    0.46
    À
    0.45
    Kl
    0.44
    ZnO
    0.44
    UE
    0.44
    Zn
    0.43
    X
    0.43
    POSITIVE LOGITS
     Duchess
    0.46
     Worth
    0.44
     geri
    0.43
     Aware
    0.43
     Northumberland
    0.42
     moments
    0.42
     Gyan
    0.42
     battered
    0.42
    0.42
     Drift
    0.41
    Act Density 0.002%

    No Known Activations