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

    Neuronpedia

    INDEX
    Explanations

    language roots (Russian, Spanish, English)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    讓人
    1.64
     relato
    1.39
     namelijk
    1.38
     كله
    1.35
     sizin
    1.29
     considérable
    1.29
     října
    1.29
     totalité
    1.27
    iędzy
    1.26
     plupart
    1.24
    POSITIVE LOGITS
    in
    2.11
    er
    1.63
    at
    1.30
    en
    1.28
    та
    1.27
    до
    1.22
    ка
    1.19
    ar
    1.14
    am
    1.14
    🏼
    1.11
    Act Density 0.020%

    No Known Activations