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

    Neuronpedia

    INDEX
    Explanations

    references to concentric structures and their geometrical properties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     endblock
    -0.36
    好
    -0.35
    extrême
    -0.33
    jš
    -0.32
    oward
    -0.31
    Checksum
    -0.31
     cuarzo
    -0.31
     Northwestern
    -0.30
    pub
    -0.30
     линей
    -0.30
    POSITIVE LOGITS
     concentric
    1.01
     circle
    0.98
     radius
    0.94
     spherical
    0.94
     Spherical
    0.93
     circles
    0.92
    circles
    0.91
    Circle
    0.91
    Circles
    0.91
     Circle
    0.91
    Act Density 0.952%

    No Known Activations