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

    Neuronpedia

    INDEX
    Explanations

    this placeholder represents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     сист
    0.95
     systemu
    0.92
     Sistem
    0.86
     sistem
    0.86
     systému
    0.85
    系统中
    0.84
     системи
    0.82
     système
    0.82
    浿
    0.82
     sistema
    0.81
    POSITIVE LOGITS
     represents
    0.98
     representing
    0.84
     representations
    0.81
     Represents
    0.78
     represent
    0.78
     representa
    0.75
     represented
    0.73
     symbolizes
    0.68
    represented
    0.64
     rappresenta
    0.64
    Act Density 0.485%

    No Known Activations