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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron is primarily firing on numeric and citation‐style tokens—e.g. years, measurements, parenthetical references—marking points in the text where data or publication details appear.

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

    sentences that express an evaluation, judgment, or other subjective/opinionated comment.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    VIN
    -0.06
     certainty
    -0.06
    979
    -0.06
    ILED
    -0.06
     ス
    -0.06
    .Marker
    -0.06
     Lifetime
    -0.06
     équip
    -0.06
    Kar
    -0.06
    IX
    -0.06
    POSITIVE LOGITS
    /react
    0.08
    -react
    0.08
    :"",
    0.07
     ArrayList
    0.07
     parenthesis
    0.06
    。この
    0.06
     nedir
    0.06
    ')↵↵
    0.06
     。
    0.06
     "../../../
    0.06
    Act Density 0.239%

    No Known Activations