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

    Neuronpedia

    INDEX
    Explanations

    boundary conditions and violations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Arrange
    0.40
     Rankings
    0.38
     mitigate
    0.38
    خة
    0.37
     Ellie
    0.37
     Ensure
    0.36
     compel
    0.36
    Arrange
    0.36
     presupuesto
    0.36
     Encourage
    0.36
    POSITIVE LOGITS
     boundaries
    0.71
     boundary
    0.67
    boundaries
    0.67
     límites
    0.61
     Boundaries
    0.61
     demarcation
    0.60
    boundary
    0.59
     Boundary
    0.57
    边界
    0.57
    Boundary
    0.56
    Act Density 0.031%

    No Known Activations