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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron doesn’t actually activate on any of the input tokens—it never fires.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    warts
    -0.07
    nictví
    -0.07
    ΑΣ
    -0.07
     školy
    -0.07
    .tags
    -0.07
    aspers
    -0.07
    節
    -0.06
     알
    -0.06
    ULATOR
    -0.06
    isempty
    -0.06
    POSITIVE LOGITS
    _su
    0.06
     conhe
    0.06
    ^
    0.06
    elerinin
    0.06
    Statement
    0.06
     Tcp
    0.05
     //(
    0.05
     advertisers
    0.05
     जम
    0.05
     Ade
    0.05
    Act Density 0.024%

    No Known Activations