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

    Neuronpedia

    INDEX
    Explanations

    linear algebra and functions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ti
    3.69
    ن
    3.40
    t
    3.40
    tan
    3.37
    ts
    3.25
    l
    3.19
    ture
    3.15
    ように
    3.10
    tela
    3.05
    tu
    3.03
    POSITIVE LOGITS
    andowski
    3.19
     liet
    3.16
     eccentricity
    3.09
    imize
    2.82
    iousness
    2.67
    ⟋
    2.64
     stature
    2.60
    ార్క్
    2.55
    ively
    2.52
     bū
    2.48
    Act Density 0.054%

    No Known Activations