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

    Neuronpedia

    INDEX
    Explanations

    "

    np_max-act · gemini-2.0-flash

    The neuron fires on uppercase identifier tokens—especially error/status codes or constant names in all‐caps.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     scraped
    -0.06
    ывать
    -0.06
    .getFullYear
    -0.06
    .ctrl
    -0.06
    ルの
    -0.06
    _wave
    -0.06
    .invalidate
    -0.06
     ulaş
    -0.06
    _alpha
    -0.06
    Narrated
    -0.06
    POSITIVE LOGITS
     Cumhuriyeti
    0.07
    Capt
    0.07
    uggestion
    0.06
    青
    0.06
     tão
    0.06
     Evil
    0.06
    Really
    0.06
    Vec
    0.06
    под
    0.06
     Dias
    0.06
    Act Density 0.037%

    No Known Activations