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

    Neuronpedia

    INDEX
    Explanations

    ,

    np_max-act · gemini-2.0-flash

    This neuron activates on LaTeX/math‐mode tokens (e.g. “$”, “\”, “^{”, “}”) and so detects formula markup.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Qu
    -0.07
    _ARGUMENT
    -0.06
    .stringify
    -0.06
     Basically
    -0.06
     Qu
    -0.06
    째
    -0.06
    Deal
    -0.06
     Âu
    -0.06
     worms
    -0.06
    oran
    -0.06
    POSITIVE LOGITS
     ucz
    0.07
    !"
    0.07
    .Has
    0.06
     Routes
    0.06
     thép
    0.06
     đáng
    0.06
     %"
    0.06
    -code
    0.06
    Tax
    0.06
    /mol
    0.06
    Act Density 0.056%

    No Known Activations