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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    This neuron detects snippets of source code by spotting programming-language keywords and structural tokens (like class, def, func, import, package, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (chip
    -0.07
     recoil
    -0.06
     Ми
    -0.06
     Jing
    -0.06
     Whip
    -0.06
    (interp
    -0.06
     rapidly
    -0.06
     bpm
    -0.06
    提出
    -0.06
    ScrollIndicator
    -0.06
    POSITIVE LOGITS
     adolescente
    0.07
    °С
    0.06
    ToObject
    0.06
    нг
    0.06
    .Serializer
    0.06
    isz
    0.06
    amon
    0.06
    대의
    0.06
    оров
    0.06
    landers
    0.06
    Act Density 0.080%

    No Known Activations