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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     arguments
    -0.08
     becoming
    -0.08
     involved
    -0.08
    ใ
    -0.08
     gas
    -0.07
     arguing
    -0.07
    เสีย
    -0.07
    人成
    -0.07
     politiek
    -0.07
    -ton
    -0.07
    POSITIVE LOGITS
    Warehouse
    0.09
     downtime
    0.09
    dbo
    0.08
     शव
    0.08
     до
    0.08
     отдела
    0.08
     অক্টোবর
    0.08
     scrapy
    0.08
    큐
    0.08
     сотрудники
    0.08
    Act Density 0.001%

    No Known Activations