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

    Neuronpedia

    INDEX
    Explanations

    hyphen

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens and mathematical expressions (numbers, decimals, and symbols in equations).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (NO
    -0.06
     arterial
    -0.06
     ав
    -0.06
    _packet
    -0.06
    ulent
    -0.06
     Pitch
    -0.06
    .П
    -0.06
     aquatic
    -0.06
    (version
    -0.06
     ит
    -0.06
    POSITIVE LOGITS
    _conn
    0.07
    ตล
    0.06
    Personally
    0.06
     looph
    0.06
     był
    0.06
     можлив
    0.06
    TASK
    0.06
     hoping
    0.06
    �
    0.06
     puerto
    0.06
    Act Density 0.006%

    No Known Activations