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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    The neuron fires on programming‐style syntax (e.g. keywords, punctuation, and identifiers), effectively detecting when the text is source code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    漫画
    -0.08
    maze
    -0.07
    placing
    -0.06
    apore
    -0.06
     Dependencies
    -0.06
     bổ
    -0.06
    .crypto
    -0.06
    _locator
    -0.06
    _stdio
    -0.06
     displ
    -0.06
    POSITIVE LOGITS
     độ
    0.07
     sand
    0.07
     Hội
    0.07
    0.07
    #,
    0.07
    خص
    0.06
    )._
    0.06
    ски
    0.06
     alg
    0.06
    =\"
    0.06
    Act Density 0.078%

    No Known Activations