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

    Neuronpedia

    INDEX
    Explanations

    code and numbers

    np_max-act · gemini-2.0-flash

    Explanation of neuron 4 behavior: the main thing this neuron does is find numeric tokens representing decimal or floating-point values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     güzel
    -0.06
     Golf
    -0.06
     Tunnel
    -0.06
     Inspection
    -0.06
    /W
    -0.06
    (players
    -0.06
    ικη
    -0.06
     cadastr
    -0.05
    世紀
    -0.05
     Tatto
    -0.05
    POSITIVE LOGITS
    frames
    0.07
    moth
    0.07
    ้าส
    0.06
     кот
    0.06
    0.06
     риз
    0.06
     strokeWidth
    0.06
    .valid
    0.06
    salary
    0.06
     vowed
    0.06
    Act Density 0.003%

    No Known Activations