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

    Neuronpedia

    INDEX
    Explanations

    affirm

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of appellate review outcomes, especially forms of “affirm” (e.g., “affirm,” “affirming,” “affirmed”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ToStr
    -0.07
     take
    -0.07
     slept
    -0.07
     Buk
    -0.07
    Range
    -0.07
     got
    -0.07
    .sys
    -0.06
    	let
    -0.06
    -move
    -0.06
     Gos
    -0.06
    POSITIVE LOGITS
     wise
    0.07
     пищ
    0.07
    _AS
    0.06
     수행
    0.06
     accountability
    0.06
    (tx
    0.06
     FactoryGirl
    0.06
    pivot
    0.06
     VStack
    0.06
    wav
    0.06
    Act Density 0.004%

    No Known Activations