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

    Neuronpedia

    INDEX
    Explanations

    code ending with semicolon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Hub
    0.41
     নিয়ম
    0.40
     CBS
    0.39
     piece
    0.39
     Sox
    0.39
     Hubbard
    0.38
     Piazza
    0.38
     Muñoz
    0.38
     Garza
    0.38
     HUB
    0.38
    POSITIVE LOGITS
    bereitung
    0.51
    mselves
    0.44
    䖝
    0.44
    uzie
    0.43
    utives
    0.42
    isiones
    0.41
    onej
    0.41
    .",
    0.41
    atient
    0.40
    зи
    0.40
    Act Density 0.003%

    No Known Activations