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

    Neuronpedia

    INDEX
    Explanations

    (

    np_max-act · gemini-2.0-flash

    The neuron never activates on any tokens, effectively detecting the absence of any targeted pattern.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    �
    -0.07
     drones
    -0.07
     plaint
    -0.07
     tart
    -0.07
    .lineWidth
    -0.07
     Attacks
    -0.07
    .RegularExpressions
    -0.06
    题
    -0.06
     triangles
    -0.06
     msgstr
    -0.06
    POSITIVE LOGITS
    ,class
    0.06
     Dell
    0.06
     (↵
    0.06
     bouts
    0.06
     meydana
    0.06
     '}↵
    0.06
    äll
    0.06
     Совет
    0.06
     Farage
    0.06
    (Call
    0.06
    Act Density 0.002%

    No Known Activations