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

    Neuronpedia

    INDEX
    Explanations

    negation and avoidance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     captions
    0.42
     variable
    0.42
     conditioner
    0.41
     downs
    0.41
     plane
    0.41
     redor
    0.41
    mlx
    0.40
     statuses
    0.40
    聽
    0.40
    पाई
    0.40
    POSITIVE LOGITS
     Мат
    0.51
     гото
    0.48
     décerné
    0.47
    уч
    0.46
     begitu
    0.46
     став
    0.45
    став
    0.45
    ну
    0.45
     අ
    0.45
     Аб
    0.44
    Act Density 0.026%

    No Known Activations