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

    Neuronpedia

    INDEX
    Explanations

    programming code

    np_max-act · gemini-2.0-flash

    The neuron is primarily activated by German-language words (e.g., Klasse, Beispiel), i.e. it detects German text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    hledem
    -0.06
     complains
    -0.06
    count
    -0.06
    shm
    -0.06
     Experience
    -0.06
    ůsob
    -0.06
     observations
    -0.06
    .WriteString
    -0.06
     дит
    -0.06
    ienza
    -0.06
    POSITIVE LOGITS
    _bi
    0.07
    Louis
    0.06
    Keep
    0.06
    收
    0.06
    RO
    0.06
     حتى
    0.06
     �
    0.06
    讓
    0.06
    .Mvc
    0.06
     рег
    0.06
    Act Density 0.033%

    No Known Activations