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

    Neuronpedia

    INDEX
    Explanations

    German

    np_max-act · gemini-2.0-flash

    The neuron activates on common German function words and filler particles, effectively signaling that the text is in German.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ENTA
    -0.07
     report
    -0.07
    couz
    -0.07
     Director
    -0.07
    con
    -0.07
     Keep
    -0.07
    icons
    -0.07
    jections
    -0.07
    _NR
    -0.06
    Element
    -0.06
    POSITIVE LOGITS
     noch
    0.07
    'util
    0.07
     niet
    0.07
     vlastně
    0.07
     jedoch
    0.06
     nur
    0.06
    -peer
    0.06
     nicht
    0.06
     вот
    0.06
     bitte
    0.06
    Act Density 0.050%

    No Known Activations