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

    Neuronpedia

    INDEX
    Explanations

    writing or text generation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     associativity
    0.48
    利用
    0.45
    actose
    0.44
     geometries
    0.42
     eigenstates
    0.40
     subspaces
    0.40
     kinases
    0.40
     unitarity
    0.40
     celebs
    0.39
     contactless
    0.39
    POSITIVE LOGITS
     written
    0.66
     prose
    0.64
     написано
    0.61
     Written
    0.59
     निबंध
    0.59
     כת
    0.58
     schreiben
    0.57
     написать
    0.57
     текст
    0.57
     Descriptions
    0.57
    Act Density 0.272%

    No Known Activations