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

    Neuronpedia

    INDEX
    Explanations

    derivative

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens that are specialized mathematical or technical terms (e.g. “map,” “finite,” “elliptic,” “limit,” “sin,” “vector,” “complex,” “derivative,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    mathematical expressions and operations related to derivatives.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Var
    -0.06
    ющего
    -0.06
     outside
    -0.06
     lbl
    -0.05
     nowadays
    -0.05
     Balls
    -0.05
    说道
    -0.05
     termin
    -0.05
     팔
    -0.05
     Athena
    -0.05
    POSITIVE LOGITS
    .LoggerFactory
    0.07
    ردد
    0.07
    <usize
    0.07
    ocê
    0.07
    _lowercase
    0.06
     ışı
    0.06
     kg
    0.06
    ры
    0.06
    เปล
    0.06
     Millionen
    0.06
    Act Density 0.108%

    No Known Activations