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

    Neuronpedia

    INDEX
    Explanations

    left right

    np_max-act · gemini-2.0-flash

    This neuron primarily detects references to laterality (e.g., “left,” “right,” “bilateral,” and similar side comparisons).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     ข
    -0.06
    mousedown
    -0.06
    cluded
    -0.06
     عليك
    -0.06
    AMPLE
    -0.06
    <Rigidbody
    -0.06
    }))↵↵
    -0.06
    _PUSHDATA
    -0.06
    illo
    -0.06
    typeof
    -0.06
    POSITIVE LOGITS
    竞
    0.07
     environ
    0.07
    245
    0.06
     Watson
    0.06
     外部リンク
    0.06
    กฎหมาย
    0.06
     ait
    0.06
     квітня
    0.06
    fas
    0.06
     wore
    0.06
    Act Density 0.008%

    No Known Activations