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

    Neuronpedia

    INDEX
    Explanations

    verified

    np_max-act · gemini-2.0-flash

    This neuron activates on the decimal‐fraction parts of numeric values (i.e. the “.###” portions of floating-point numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _spectrum
    -0.07
     acquisition
    -0.07
     compatibility
    -0.07
     Partners
    -0.06
    $text
    -0.06
     acceleration
    -0.06
     analytics
    -0.06
     опер
    -0.06
     jazz
    -0.06
     supporters
    -0.06
    POSITIVE LOGITS
     alespoň
    0.07
    ワー
    0.07
    InstanceOf
    0.06
     سریع
    0.06
    怖
    0.06
    _DONE
    0.06
    IVO
    0.06
    _social
    0.06
    로운
    0.06
     соци
    0.06
    Act Density 0.171%

    No Known Activations