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 occurrences of “filename” (e.g. struct fields or variables named filename) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Holder
    -0.07
    .Done
    -0.07
     Java
    -0.06
     root
    -0.06
     한
    -0.06
     vegetation
    -0.06
    Viewport
    -0.06
     takový
    -0.06
     ECC
    -0.06
     Dit
    -0.06
    POSITIVE LOGITS
    _session
    0.07
    ()].
    0.06
     erót
    0.06
    agment
    0.06
    ьогод
    0.06
    alchemy
    0.06
    paren
    0.06
    .getTransaction
    0.06
    getSize
    0.06
    西省
    0.06
    Act Density 0.246%

    No Known Activations