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

    Neuronpedia

    INDEX
    Explanations

    cuda compilation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    िंग्स
    1.11
    um
    1.10
    ut
    1.09
    ’
    1.08
    ètes
    1.04
    ęła
    1.02
    ه‌ای
    1.01
     апре
    0.99
    ittances
    0.99
    ru
    0.98
    POSITIVE LOGITS
    ك
    1.38
    ↵
    1.32
    c
    1.28
    ี
    1.21
    тся
    1.13
    то
    1.07
    v
    1.07
    x
    1.02
    al
    1.00
     critic
    1.00
    Act Density 0.000%

    No Known Activations