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

    Neuronpedia

    INDEX
    Explanations

    offer help or ask for clarification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     belongs
    0.92
     embraces
    0.75
    collabor
    0.74
    Its
    0.71
     belong
    0.70
     принадлежит
    0.70
     adopts
    0.68
    It
    0.68
     collaborate
    0.67
     piloted
    0.67
    POSITIVE LOGITS
     maybe
    1.02
     magari
    0.97
     quizás
    0.97
     perhaps
    0.95
     yoki
    0.91
     Dacă
    0.91
     puedes
    0.90
     Puoi
    0.89
     posiblemente
    0.89
     quizá
    0.89
    Act Density 0.080%

    No Known Activations