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

    Neuronpedia

    INDEX
    Explanations

    hypothetical or concept components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     приме
    1.10
    वटी
    1.05
     sửa
    0.99
     pamoja
    0.98
    ලි
    0.97
     ⿰
    0.97
    ಇದ
    0.96
    運用
    0.96
    𝗳
    0.96
     bords
    0.95
    POSITIVE LOGITS
     beloved
    1.04
     paradigma
    1.03
     Beloved
    1.03
     exiting
    1.02
    假设
    1.02
    ください
    1.00
    当然
    0.99
     régimen
    0.98
     octane
    0.98
    显然
    0.97
    Act Density 0.000%

    No Known Activations