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

    Neuronpedia

    INDEX
    Explanations

    acting as personas and roles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    risk
    0.48
    census
    0.46
    tolerance
    0.46
     риска
    0.45
    measure
    0.44
    nums
    0.43
    篩
    0.42
    inclusion
    0.42
     tolerances
    0.42
    solutions
    0.41
    POSITIVE LOGITS
     말투
    0.92
     imperson
    0.89
     simulating
    0.87
     simulate
    0.85
     simulated
    0.84
    模拟
    0.80
     personaje
    0.78
     Simulate
    0.77
     personagem
    0.77
     simulates
    0.77
    Act Density 0.493%

    No Known Activations