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

    Neuronpedia

    INDEX
    Explanations

    loop iteration with i

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     of
    -1.05
     our
    -1.02
    っていたので
    -0.95
    esía
    -0.91
     even
    -0.90
    Архівовано
    -0.90
     allí
    -0.88
    fillColor
    -0.86
    AHUL
    -0.85
     députés
    -0.84
    POSITIVE LOGITS
     individuellen
    0.89
    Bibliografie
    0.88
    oodles
    0.82
     spezi
    0.81
     takich
    0.81
    wiches
    0.78
    瞠
    0.77
     român
    0.77
    度に
    0.77
    ramento
    0.76
    Act Density 0.001%

    No Known Activations