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

    Neuronpedia

    INDEX
    Explanations

    guard, vigilance, and protection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )
    0.86
    s
    0.70
    ne
    0.68
    d
    0.66
    ir
    0.66
    na
    0.66
    in
    0.65
    .")
    0.64
     computeEncoder
    0.63
    u
    0.62
    POSITIVE LOGITS
     guard
    0.96
     guards
    0.92
     guarding
    0.86
    guards
    0.86
     vigilance
    0.82
     Guards
    0.80
     Guard
    0.79
     охра
    0.78
     vigilancia
    0.78
     patrol
    0.72
    Act Density 0.022%

    No Known Activations