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

    Neuronpedia

    INDEX
    Explanations

    beginning of segments or concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    bo
    0.78
    Til
    0.77
     св
    0.74
    ինչ
    0.74
    ěž
    0.74
     marred
    0.73
     Til
    0.70
    covering
    0.70
     Erosion
    0.69
    因为它
    0.68
    POSITIVE LOGITS
     minuto
    0.96
    WICK
    0.88
     elementi
    0.86
     Wyman
    0.85
     Rory
    0.84
     kwal
    0.82
     nueva
    0.82
     nuevo
    0.81
     piatti
    0.80
     utenti
    0.80
    Act Density 0.000%

    No Known Activations