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

    Neuronpedia

    INDEX
    Explanations

    indicates a temporal sequence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.32
    の
    0.28
    ۔
    0.28
     يقول
    0.27
     WHEN
    0.27
     badań
    0.26
     ngunit
    0.26
     ketika
    0.26
     sırasında
    0.26
     إذا
    0.25
    POSITIVE LOGITS
    being
    0.28
     being
    0.26
    cura
    0.25
    cuss
    0.24
    curr
    0.24
    chy
    0.24
    dept
    0.24
    ↵
    0.24
    ório
    0.24
    t
    0.23
    Act Density 0.102%

    No Known Activations