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

    Neuronpedia

    INDEX
    Explanations

    20

    np_max-act · gemini-2.0-flash

    the neuron detects numeric information—numbers, years, percentages and other digit-based statistics in the text.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    ajan
    -0.07
    flammatory
    -0.06
    inski
    -0.06
    _city
    -0.06
     ago
    -0.06
     lieu
    -0.06
     purse
    -0.06
     getLast
    -0.06
    �
    -0.06
    holder
    -0.06
    POSITIVE LOGITS
    [arr
    0.07
    (nome
    0.06
     wb
    0.06
     Relatives
    0.06
    etermination
    0.06
     Clay
    0.06
    还
    0.06
     acknowledgment
    0.06
    ;margin
    0.06
     QT
    0.06
    Act Density 0.643%

    No Known Activations