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

    Neuronpedia

    INDEX
    Explanations

    absence or departure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    শে
    -1.09
     dokonce
    -0.89
    สวย
    -0.87
    ꪀ
    -0.86
    kách
    -0.86
    Saharan
    -0.85
     mahogany
    -0.85
     sapatos
    -0.84
    rometers
    -0.84
    ництво
    -0.83
    POSITIVE LOGITS
     recent
    1.09
     ihres
    1.02
    ñadir
    1.01
     muszą
    0.99
    みたいに
    0.98
     отсутствие
    0.94
    several
    0.91
     biens
    0.91
     THREE
    0.90
     leader
    0.90
    Act Density 0.035%

    No Known Activations