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

    Neuronpedia

    INDEX
    Explanations

    what is

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    When
    1.11
     когато
    0.98
    While
    0.98
    Although
    0.97
    Whether
    0.96
    How
    0.96
    Cuando
    0.92
    Him
    0.91
    Since
    0.90
    にとって
    0.90
    POSITIVE LOGITS
     constitutes
    2.13
     exists
    2.03
     they
    1.95
     underlies
    1.89
     happens
    1.88
     awaits
    1.85
     precedes
    1.78
     motivates
    1.75
     comes
    1.75
     dominates
    1.74
    Act Density 0.176%

    No Known Activations