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

    Neuronpedia

    INDEX
    Explanations

    Case

    np_max-act · gemini-2.0-flash

    This neuron detects headings that label court cases—specifically tokens in the “Appellate Case” and “Trial Court Case” lines.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    iami
    -0.07
    occan
    -0.06
     brows
    -0.06
    .touch
    -0.06
     Regents
    -0.06
     Worship
    -0.06
     مهد
    -0.06
     اخت
    -0.06
    รร
    -0.06
    нюю
    -0.06
    POSITIVE LOGITS
    Preferred
    0.07
     UM
    0.06
    ��
    0.06
    "All
    0.06
    20
    0.06
    _armor
    0.06
    Alle
    0.06
     jorn
    0.06
    	InputStream
    0.06
    ΗΜΑ
    0.06
    Act Density 0.003%

    No Known Activations