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

    Neuronpedia

    INDEX
    Explanations

    perform

    np_max-act · gemini-2.0-flash

    The neuron primarily responds to numeric tokens (e.g. integers or floating-point numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cluster
    -0.07
     clusters
    -0.07
    _exclude
    -0.07
     gu
    -0.07
    ACKET
    -0.07
    kuk
    -0.07
     अगस
    -0.06
     Mutable
    -0.06
    about
    -0.06
    _subscription
    -0.06
    POSITIVE LOGITS
     perform
    0.11
     Perform
    0.10
     performed
    0.10
     Performs
    0.09
     performs
    0.09
     수행
    0.09
    Perform
    0.08
     performing
    0.08
    を行
    0.08
    رم
    0.07
    Act Density 0.029%

    No Known Activations