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

    Neuronpedia

    INDEX
    Explanations

    problem potential roadblocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     haute
    0.50
     reproductions
    0.50
     splashes
    0.50
     gastronomy
    0.49
     ustaw
    0.48
     fabricants
    0.48
     monaster
    0.47
     reborn
    0.47
     služ
    0.47
     przechowy
    0.47
    POSITIVE LOGITS
    决策
    0.80
     Decision
    0.80
    Decision
    0.78
     decision
    0.76
     Decisions
    0.69
     рішення
    0.68
     সিদ্ধান্ত
    0.68
    decision
    0.67
     решение
    0.64
     decisión
    0.64
    Act Density 0.314%

    No Known Activations