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

    Neuronpedia

    INDEX
    Explanations

    mirror

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     stakes
    -0.07
     heap
    -0.07
     displacement
    -0.07
     selection
    -0.07
     oxidative
    -0.06
     collect
    -0.06
     wię
    -0.06
     Income
    -0.06
    783
    -0.06
     punk
    -0.06
    POSITIVE LOGITS
     mirrors
    0.11
     Mirror
    0.11
     mirror
    0.10
    Mirror
    0.08
    mirror
    0.08
    러
    0.08
     mirrored
    0.07
    र
    0.07
    镜
    0.07
    鏡
    0.07
    Act Density 0.004%

    No Known Activations