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

    Neuronpedia

    INDEX
    Explanations

    levels of detail

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     намного
    1.01
    不仅
    0.93
     EVERY
    0.93
    EVERY
    0.92
     ogrom
    0.91
     EVERYTHING
    0.89
     emberek
    0.89
     wielu
    0.88
     огром
    0.86
     гораздо
    0.86
    POSITIVE LOGITS
     so
    0.98
     slightly
    0.98
     hopefully
    0.90
     suggestions
    0.89
     depending
    0.88
    以便
    0.87
     feel
    0.86
     respectively
    0.85
    方便
    0.84
     slight
    0.83
    Act Density 0.727%

    No Known Activations