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

    Neuronpedia

    INDEX
    Explanations

    preserving

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Watches
    -0.08
    afone
    -0.08
    TEAM
    -0.07
    631
    -0.06
    819
    -0.06
    /**<
    -0.06
    Means
    -0.06
    ('{}
    -0.06
    われ
    -0.06
     RO
    -0.06
    POSITIVE LOGITS
    preserve
    0.07
    正确
    0.06
     detailed
    0.06
     aug
    0.06
     environments
    0.06
     Comment
    0.06
    ением
    0.06
     有
    0.06
     влия
    0.06
     olmasına
    0.06
    Act Density 0.022%

    No Known Activations