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

    Neuronpedia

    INDEX
    Explanations

    evaluate

    np_max-act · gemini-2.0-flash

    The neuron responds to the occurrence of an evaluation instruction—specifically spotting the word “evaluate” (often as part of a directive to assess something).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    notation
    -0.07
    loy
    -0.06
    659
    -0.06
     induction
    -0.06
    Inverse
    -0.06
     occurrence
    -0.06
    loo
    -0.06
    ultipart
    -0.06
    (surface
    -0.06
    KK
    -0.06
    POSITIVE LOGITS
    рахов
    0.07
    PointF
    0.07
     ।↵
    0.07
    -root
    0.06
     Start
    0.06
    اء
    0.06
    _total
    0.06
    _spot
    0.06
    به
    0.06
    Themes
    0.06
    Act Density 0.003%

    No Known Activations