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

    Neuronpedia

    INDEX
    Explanations

    technical labels

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Did
    1.09
    took
    1.07
    did
    1.01
    didn
    0.95
    Have
    0.93
    am
    0.85
    have
    0.85
    intra
    0.82
    take
    0.81
     did
    0.79
    POSITIVE LOGITS
     Really
    0.85
     WOULD
    0.84
     worries
    0.83
     ".
    0.82
     Mejor
    0.82
     I
    0.78
     Specifications
    0.78
     размеры
    0.78
     autres
    0.77
     cartas
    0.77
    Act Density 0.000%

    No Known Activations