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

    Neuronpedia

    INDEX
    Explanations

    Here's a rule/recipe/example

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    roskop
    0.36
     franchisees
    0.32
     output
    0.32
     disrupted
    0.32
     sometime
    0.32
     mutual
    0.32
     grease
    0.31
     бү
    0.31
    ́ng
    0.31
     دیں
    0.31
    POSITIVE LOGITS
    Product
    0.96
     Product
    0.92
     product
    0.87
    product
    0.80
     Produkt
    0.79
     produto
    0.79
     produkt
    0.75
    产品
    0.75
     prodotto
    0.73
     producto
    0.72
    Act Density 0.001%

    No Known Activations