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

    Neuronpedia

    INDEX
    Explanations

    iteration

    np_max-act · gemini-2.0-flash

    This neuron detects LaTeX summation limit markers (e.g. “limits” and the associated subscript/superscript notation) in math expressions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    okemon
    -0.07
     Gar
    -0.07
    चन
    -0.06
     degraded
    -0.06
    clearfix
    -0.06
     "}
    -0.06
     azure
    -0.06
     metropolitan
    -0.06
     _.
    -0.06
    expand
    -0.06
    POSITIVE LOGITS
    ejména
    0.08
    QUAL
    0.08
    (Q
    0.06
     ủy
    0.06
     عملیات
    0.06
    >\↵
    0.06
     wanting
    0.06
    \Action
    0.06
    								↵
    0.06
     อำ
    0.06
    Act Density 0.008%

    No Known Activations