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

    Neuronpedia

    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    This neuron detects numeric tokens representing the fractional (decimal) part of numbers (i.e. the “.xxx” in floating-point numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    /bootstrap
    -0.07
    847
    -0.07
    -ts
    -0.07
     Article
    -0.06
    /info
    -0.06
    -vers
    -0.06
     để
    -0.06
     Jul
    -0.06
    .And
    -0.06
    _Version
    -0.06
    POSITIVE LOGITS
    ，本
    0.06
     MV
    0.06
     CHILD
    0.06
     наш
    0.06
     нес
    0.06
     rekl
    0.06
    рач
    0.06
    átek
    0.06
     진행
    0.06
     vous
    0.06
    Act Density 0.135%

    No Known Activations