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

    Neuronpedia

    INDEX
    Explanations

    close and reopen

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Polar
    0.84
    il
    0.83
     Initialization
    0.78
     Initial
    0.76
     Profit
    0.75
     Verification
    0.72
     Utilization
    0.72
     Polarization
    0.72
     Pressure
    0.72
     Direction
    0.72
    POSITIVE LOGITS
    ה
    0.95
     경우에는
    0.84
    ק
    0.82
    ه
    0.81
    يا
    0.77
     offices
    0.75
     एचडी
    0.73
    آ
    0.73
     pled
    0.73
    ADE
    0.71
    Act Density 0.001%

    No Known Activations