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

    Neuronpedia

    INDEX
    Explanations

    solving equations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     よる
    1.16
     voglia
    1.15
    rijven
    1.10
     hilly
    1.10
     unifying
    1.08
     ciudadanos
    1.08
     pamoja
    1.08
     jugo
    1.08
     cómod
    1.08
     chuy
    1.08
    POSITIVE LOGITS
    다른
    1.04
    दी
    1.00
    sonsten
    1.00
    그러나
    1.00
    すべての
    1.00
    ೇಶ
    0.98
    ઓના
    0.96
     다른
    0.95
    Former
    0.93
    ът
    0.93
    Act Density 0.003%

    No Known Activations