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

    Neuronpedia

    INDEX
    Explanations

    transformer and attention

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     TRAINING
    0.45
     Trainers
    0.44
     dugout
    0.44
     sentences
    0.43
     трени
    0.42
     Entrenamiento
    0.42
    훈련
    0.42
     Trainings
    0.42
     سیاه
    0.42
    لود
    0.41
    POSITIVE LOGITS
     attention
    0.86
    attention
    0.80
    Attention
    0.80
    注意力
    0.78
     Attention
    0.75
     внимание
    0.68
     ધ્યાન
    0.65
     attentions
    0.62
    transformer
    0.62
     внимания
    0.61
    Act Density 0.057%

    No Known Activations