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

    Neuronpedia

    INDEX
    Explanations

    attenu

    np_max-act · gemini-2.0-flash

    The neuron detects verbs (and related forms) that describe therapeutic or inhibitory actions—words like “suppresses,” “attenuates,” “inhibits,” “prevents,” etc., signaling reduction of a pathological process.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     teil
    -0.07
     цен
    -0.06
    .Assign
    -0.06
    880
    -0.06
     imprisonment
    -0.06
     поруш
    -0.06
    hes
    -0.06
    urtle
    -0.06
    _deep
    -0.06
     túi
    -0.06
    POSITIVE LOGITS
     ABC
    0.07
     difficulty
    0.07
     evidently
    0.07
     Unc
    0.07
    scriptions
    0.06
    има
    0.06
    -coordinate
    0.06
     Cord
    0.06
     ORD
    0.06
    -associated
    0.06
    Act Density 0.039%

    No Known Activations