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

    Neuronpedia

    INDEX
    Explanations

    P

    np_max-act · gemini-2.0-flash

    The neuron fires on the “P” token in the “NOT FOR PUBLICATION” header, i.e. it detects the document’s non-publication label.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    legal terminology related to immigration cases and appeals.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     whats
    -0.07
    odes
    -0.07
    หา
    -0.06
    dyby
    -0.06
    utters
    -0.06
    veillance
    -0.06
     nuclei
    -0.06
    hm
    -0.06
    toBe
    -0.06
    Preview
    -0.06
    POSITIVE LOGITS
     creeping
    0.07
    ετ
    0.07
    (datos
    0.06
     جد
    0.06
    -angular
    0.06
     hijo
    0.06
     рес
    0.06
     dishonest
    0.06
     Mandarin
    0.06
    طب
    0.06
    Act Density 0.005%

    No Known Activations