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

    Neuronpedia

    INDEX
    Explanations

    articles

    np_max-act · gemini-2.0-flash

    The neuron detects the start-of-text token (i.e., the very beginning of a document).

    oai_token-act-pair · gpt-5-miniTriggered by @tatsatx
    New Auto-Interp
    Negative Logits
    ).
    -0.08
     съ
    -0.08
     gcd
    -0.08
     arithmetic
    -0.07
     πι
    -0.07
     symmetrical
    -0.07
     mathematic
    -0.07
     neg
    -0.07
    Arithmetic
    -0.07
     sayı
    -0.07
    POSITIVE LOGITS
    <_
    0.08
     PTSD
    0.08
     curator
    0.08
     curated
    0.08
     organizar
    0.08
     curate
    0.08
     Salon
    0.08
     fronte
    0.08
     especializado
    0.07
     beurt
    0.07
    Act Density 0.099%

    No Known Activations