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

    Neuronpedia

    INDEX
    Explanations

    div

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens—especially digits and numeric literals (including decimal fractions) often appearing in XPaths or index values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    чи
    -0.07
     unary
    -0.06
     �
    -0.06
    Iron
    -0.06
    .app
    -0.06
    "With
    -0.06
     massac
    -0.06
    َأ
    -0.05
     Halifax
    -0.05
     plight
    -0.05
    POSITIVE LOGITS
    -bre
    0.07
     snake
    0.07
    bonus
    0.07
     ngang
    0.06
    aternion
    0.06
     laugh
    0.06
    _PR
    0.06
     LETTER
    0.06
     }>↵
    0.06
     sunglasses
    0.06
    Act Density 0.004%

    No Known Activations