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

    Neuronpedia

    INDEX
    Explanations

    separates items with articles/nouns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    あった
    0.46
    Relationships
    0.45
    relationships
    0.44
    Relationship
    0.43
    relationship
    0.43
     είχε
    0.43
    關係
    0.42
    當時
    0.41
     también
    0.41
     relationship
    0.41
    POSITIVE LOGITS
     Separ
    0.52
     separating
    0.50
    separated
    0.48
    用い
    0.48
     separated
    0.48
     separa
    0.47
     séparation
    0.47
    Separator
    0.46
     newline
    0.46
    separ
    0.45
    Act Density 0.014%

    No Known Activations