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

    Neuronpedia

    INDEX
    Explanations

    Scientific/research texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Hello
    -0.07
    进而
    -0.07
    🦄
    -0.07
     passwd
    -0.06
     gentlemen
    -0.06
    �
    -0.06
    識
    -0.06
     Je
    -0.06
     �
    -0.06
     speed
    -0.06
    POSITIVE LOGITS
    .Evaluate
    0.08
     Flying
    0.07
     אח
    0.07
    _MOUSE
    0.07
    𝕋
    0.07
    ﯵ
    0.06
    relevant
    0.06
    .Collection
    0.06
     spared
    0.06
    ;;;;;;;;
    0.06
    Act Density 0.051%

    No Known Activations