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

    Neuronpedia

    INDEX
    Explanations

    a few single document

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     alive
    0.17
     our
    0.17
     any
    0.16
     achieving
    0.16
     when
    0.16
     samma
    0.16
     avendo
    0.16
     attaining
    0.16
     étant
    0.16
     equaling
    0.16
    POSITIVE LOGITS
    רי
    0.18
    ד
    0.16
     DSLR
    0.15
    לי
    0.15
    0.15
     к
    0.14
     k
    0.14
     makeshift
    0.14
     ס
    0.14
     pequeño
    0.14
    Act Density 0.928%

    No Known Activations