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

    Neuronpedia

    INDEX
    Explanations

    .

    np_max-act · gemini-2.0-flash

    This neuron detects tokens with unusually low model confidence (i.e. rare or surprising tokens).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Cache
    -0.07
    411
    -0.06
    �어
    -0.06
     Rewrite
    -0.06
    elastic
    -0.06
     Stefan
    -0.06
    �
    -0.06
    conditionally
    -0.06
    .WriteString
    -0.06
     BigNumber
    -0.06
    POSITIVE LOGITS
     ustanov
    0.07
     kriz
    0.06
    iks
    0.06
    しかし
    0.06
     `[
    0.06
    :";↵
    0.06
    uco
    0.06
     مثلا
    0.06
    %;↵
    0.06
    [${
    0.06
    Act Density 0.052%

    No Known Activations