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

    Neuronpedia

    INDEX
    Explanations

    Math

    np_max-act · gemini-2.0-flash

    This neuron detects LaTeX‐style mathematical notation, especially backslash commands and dollar‐delimited formulas.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Polo
    -0.07
    .surface
    -0.07
     manager
    -0.07
    ома
    -0.06
    илися
    -0.06
    巡
    -0.06
     rebut
    -0.06
     ePub
    -0.06
     midfield
    -0.06
     месяцев
    -0.06
    POSITIVE LOGITS
    ScreenState
    0.07
    >"
    0.07
    яет
    0.07
     vự
    0.06
     objetivo
    0.06
    >";↵
    0.06
    الات
    0.06
    iliği
    0.06
     společnost
    0.06
    =subprocess
    0.06
    Act Density 0.043%

    No Known Activations