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

    Neuronpedia

    INDEX
    Explanations

    leak

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     training
    -0.07
    88
    -0.07
    เพ
    -0.06
     phase
    -0.06
    bum
    -0.06
    riger
    -0.06
    .writer
    -0.06
    _RO
    -0.06
     Tem
    -0.06
    言って
    -0.06
    POSITIVE LOGITS
    uai
    0.07
     кури
    0.07
     newPath
    0.07
     cran
    0.06
     Petr
    0.06
     energie
    0.06
     productName
    0.06
    OURCE
    0.06
    $action
    0.06
     longevity
    0.06
    Act Density 0.014%

    No Known Activations