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

    Neuronpedia

    INDEX
    Explanations

    say the ending -ators

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ethn
    -0.08
    趕
    -0.07
     gep
    -0.07
     él
    -0.07
    emap
    -0.07
    iman
    -0.07
    مارس
    -0.07
     heute
    -0.07
     Bake
    -0.07
     CAM
    -0.07
    POSITIVE LOGITS
     reliability
    0.07
    ސ
    0.07
    $res
    0.06
    ])
↵
    0.06
     slots
    0.06
     instances
    0.06
    됐
    0.06
    丁
    0.06
    ])↵
    0.06
    켤
    0.06
    Act Density 0.004%

    No Known Activations